abstract
stringlengths 383
2.62k
| TLDR
stringlengths 4
250
|
|---|---|
Diffusion models have been recently employed to improve certified robustness through the process of denoising. However, the theoretical understanding of why diffusion models are able to improve the certified robustness is still lacking, preventing from further improvement. In this study, we close this gap by analyzing the fundamental properties of diffusion models and establishing the conditions under which they can enhance certified robustness. This deeper understanding allows us to propose a new method DensePure, designed to improve the certified robustness of a pretrained model (i.e. classifier). Given an (adversarial) input, DensePure consists of multiple runs of denoising via the reverse process of the diffusion model (with different random seeds) to get multiple reversed samples, which are then passed through the classifier, followed by majority voting of inferred labels to make the final prediction. This design of using multiple runs of denoising is informed by our theoretical analysis of the conditional distribution of the reversed sample. Specifically, when the data density of a clean sample is high, its conditional density under the reverse process in a diffusion model is also high; thus sampling from the latter conditional distribution can purify the adversarial example and return the corresponding clean sample with a high probability. By using the highest density point in the conditional distribution as the reversed sample, we identify the robust region of a given instance under the diffusion model's reverse process. We show that this robust region is a union of multiple convex sets, and is potentially much larger than the robust regions identified in previous works. In practice, DensePure can approximate the label of the high density region in the conditional distribution so that it can enhance certified robustness. We conduct extensive experiments to demonstrate the effectiveness of DensePure by evaluating its certified robustness given a standard model via randomized smoothing. We show that DensePure is consistently better than existing methods on ImageNet, with 7% improvement on average. | We theoretically analyze the fundamental properties of diffusion models to understand why and how it enhances certified robustness. Inspired by the analysis, we propose a new method to improve the certified robustness of the clearn classifier |
In deep learning, fine-grained N:M sparsity reduces the data footprint and bandwidth of a General Matrix multiply (GEMM) up to x2, and doubles throughput by skipping computation of zero values. So far, it was mainly only used to prune weights to accelerate the forward and backward phases. We examine how this method can be used also for the neural gradients (i.e. loss gradients with respect to the intermediate neural layer outputs). To this end, we first establish a tensor-level optimality criteria. Previous works aimed to minimize the mean-square-error (MSE) of each pruned block. We show that while minimization of the MSE works fine for pruning the weights and activations, it catastrophically fails for the neural gradients. Instead, we show that accurate pruning of the neural gradients requires an unbiased minimum-variance pruning mask. We design such specialized masks, and find that in most cases, 1:2 sparsity is sufficient for training, and 2:4 sparsity is usually enough when this is not the case. Further, we suggest combining several such methods together in order to potentially speed up training even more. A reference implementation is supplied in the supplementary material. | A method to use structured N:M sparsity on all training GEMM operations |
Existing analysis of AdaGrad and other adaptive methods for smooth convex optimization is typically for functions with bounded domain diameter. In unconstrained problems, previous works guarantee an asymptotic convergence rate without an explicit constant factor that holds true for the entire function class. Furthermore, in the stochastic setting, only a modified version of AdaGrad, different from the one commonly used in practice, in which the latest gradient is not used to update the stepsize, has been analyzed. Our paper aims at bridging these gaps and developing a deeper understanding of AdaGrad and its variants in the standard setting of smooth convex functions as well as the more general setting of quasar convex functions. First, we demonstrate new techniques to explicitly bound the convergence rate of the vanilla AdaGrad for unconstrained problems in both deterministic and stochastic settings. Second, we propose a variant of AdaGrad for which we can show the convergence of the last iterate, instead of the average iterate. Finally, we give new accelerated adaptive algorithms and their convergence guarantee in the deterministic setting with explicit dependency on the problem parameters, improving upon the asymptotic rate shown in previous works. | New techniques to prove the convergence rate of AdaGrad and new accelerated adaptive algorithms without bounded domain assumption beyond standard convex and smooth functions |
Computational drug repurposing is a cost- and time-efficient method to identify new indications of approved or experimental drugs/compounds. It is especially critical for emerging and/or orphan diseases due to its cheaper investment and shorter research cycle compared with traditional wet-lab drug discovery approaches. However, the underlying mechanisms of action between repurposed drugs and their target diseases remain largely unknown, which is still an unsolved issue in existing repurposing methods. As such, computational drug repurposing has not been widely adopted in clinical settings. In this work, based on a massive biomedical knowledge graph, we propose a computational drug repurposing framework that not only predicts the treatment probabilities between drugs and diseases but also predicts the path-based, testable mechanisms of action (MOAs) as their biomedical explanations. Specifically, we utilize the GraphSAGE model in an unsupervised manner to integrate each entity’s neighborhood information and employ a Random Forest model to predict the treatment probabilities between pairs of drugs and diseases. Moreover, we train an adversarial actor-critic reinforcement learning model to predict the potential MOA for explaining drug purposing. To encourage the model to find biologically reasonable paths, we utilize the curated molecular interactions of drugs and a PubMed-publication-based concept distance to extract potential drug MOA paths from the knowledge graph as ”demonstration paths” to guide the model during the process of path-finding. Comprehensive experiments and case studies show that the proposed framework outperforms state-of-the-art baselines in both predictive performance of drug repurposing and explanatory performance of recapitulating human-curated DrugMechDB-based paths. | We predict drug repurposing candidates and their path-based mechanisms of action based on a large biomedical knowledge graph. |
Controlling artificial agents from visual sensory data is an arduous task. Reinforcement learning (RL) algorithms can succeed in this but require large amounts of interactions between the agent and the environment. To alleviate the issue, unsupervised RL proposes to employ self-supervised interaction and learning, for adapting faster to future tasks. Yet, whether current unsupervised strategies improve generalization capabilities is still unclear, especially in visual control settings. In this work, we design an unsupervised RL strategy for data-efficient visual control. First, we show that world models pre-trained with data collected using unsupervised RL can facilitate adaptation for future tasks. Then, we analyze several design choices to adapt faster, effectively reusing the agents' pre-trained components, and planning in imagination, with our hybrid planner, which we dub Dyna-MPC. By combining the findings of a large-scale empirical study, we establish an approach that strongly improves performance on the Unsupervised RL Benchmark, requiring 20$\times$ less data to match the performance of supervised methods. The approach also demonstrates robust performance on the Real-Word RL benchmark, hinting that the approach generalizes to noisy environments. | We combine the findings of a large-scale study on the Unsupervised RL Benchmark and propose a new hybrid planner, to establish an end-to-end approach that can be efficiently fine-tuned after unsupervised pre-training. |
Recent advances in vision-language pre-training have pushed the state-of-the-art on various vision-language tasks, making machines more capable of multi-modal writing (image-to-text generation) and painting (text-to-image generation). However, few studies investigate if these two essential capabilities can be learned together and boost each other, making a versatile and powerful multi-modal foundation model. In this work, we disclose the potential of symmetric generative vision-language pre-training in learning to write and paint concurrently, and propose a new unified modal model, named DaVinci, trained with prefix language modeling and prefix image modeling, a simple generative self-supervised objective on image-text pairs. Thanks to the proposed prefix multi-modal modeling framework, DaVinci is simple to train, scalable to huge data, adaptable to both writing and painting tasks, and also strong on other vision, text, and multi-modal understanding tasks. DaVinci achieves competitive performance on a wide range of 27 generation/understanding tasks and demonstrates the superiority of combining vision/language generative pre-training. Furthermore, we carefully benchmark the performance of different vision-language pre-training objectives on different scales of pre-training datasets on a heterogeneous and broad distribution coverage. Our results demonstrate the potential of exploiting self-supervision in both language and vision inputs, and establish new, stronger baselines for future comparisons at different data scales. The code and pre-trained models are available at https://github.com/shizhediao/DaVinci. | The paper proposes a simple, scalable, and versatile seq2seq foundation model, which is capable of vision-language understanding, image-to-text generation, and text-to-image generation with a single unified architecture |
Partial label learning (PLL) is a typical weakly supervised learning problem, where each training example is associated with a set of candidate labels among which only one is true. Most existing PLL approaches assume that the incorrect labels in each training example are randomly picked as the candidate labels and model the generation process of the candidate labels in a simple way. However, these approaches usually do not perform as well as expected due to the fact that the generation process of the candidate labels is always instance-dependent. Therefore, it deserves to be modeled in a refined way. In this paper, we consider instance-dependent PLL and assume that the generation process of the candidate labels could decompose into two sequential parts, where the correct label emerges first in the mind of the annotator but then the incorrect labels related to the feature are also selected with the correct label as candidate labels due to uncertainty of labeling. Motivated by this consideration, we propose a novel PLL method that performs Maximum A Posterior(MAP) based on an explicitly modeled generation process of candidate labels via decomposed probability distribution models. Extensive experiments on manually corrupted benchmark datasets and real-world datasets validate the effectiveness of the proposed method. | We consider instance-dependent PLL and assume that the generation process of the candidate labels could decompose into two sequential parts. |
The key to high-level cognition is believed to be the ability to systematically manipulate and compose knowledge pieces. While token-like structured knowledge representations are naturally provided in text, it is elusive how to obtain them for unstructured modalities such as scene images. In this paper, we propose a neural mechanism called Neural Systematic Binder or SysBinder for constructing a novel structured representation called Block-Slot Representation. In Block-Slot Representation, object-centric representations known as slots are constructed by composing a set of independent factor representations called blocks, to facilitate systematic generalization. SysBinder obtains this structure in an unsupervised way by alternatingly applying two different binding principles: spatial binding for spatial modularity across the full scene and factor binding for factor modularity within an object. SysBinder is a simple, deterministic, and general-purpose layer that can be applied as a drop-in module in any arbitrary neural network and on any modality. In experiments, we find that SysBinder provides significantly better factor disentanglement within the slots than the conventional object-centric methods, including, for the first time, in visually complex scene images such as CLEVR-Tex. Furthermore, we demonstrate factor-level systematicity in controlled scene generation by decoding unseen factor combinations. | We propose a novel object-centric representation called block-slots, which unlike the conventional slots, provides within-slot disentanglement via vector-formed factor representations. |
Federated learning (FL) enables learning from decentralized privacy-sensitive data, with computations on raw data confined to take place at edge clients. This paper introduces mixed FL, which incorporates an additional loss term calculated at the coordinating server (while maintaining FL’s private data restrictions). For example, additional datacenter data can be leveraged to jointly learn from centralized (datacenter) and decentralized (federated) training data and better match an expected inference data distribution.Mixed FL also enables offloading some intensive computations (e.g., embedding regularization) to the server, greatly reducing communication and client computation load. For these and other mixed FL use cases, we present three algorithms: PARALLEL TRAINING, 1-WAY GRADIENT TRANSFER, and 2-WAY GRADIENT TRANSFER. We perform extensive experiments of the algorithms on three tasks, demonstrating that mixed FL can blend training data to achieve an oracle’s accuracy on an inference distribution, and can reduce communication and computation overhead by more than 90%. Finally, we state convergence bounds for all algorithms, and give intuition on the mixed FL problems best suited to each. The theory confirms our empirical observations of how the algorithms perform under different mixed FL problem settings. | Federated learning (FL) is good (better privacy, higher accuracy), and 'mixed FL' (concurrent joint FL and centralized learning) can make it even better, by mitigating distribution shifts and saving bandwidth and compute. |
Clustering based on the random walk operator has been proven effective for undirected graphs, but its generalization to directed graphs (digraphs) is much more challenging. Although the random walk operator is well-defined for digraphs, in most cases such graphs are not strongly connected, and hence the associated random walks are not irreducible, which is a crucial property for clustering that exists naturally in the undirected setting. To remedy this, the usual workaround is to either naively symmetrize the adjacency matrix or to replace the natural random walk operator by the teleporting random walk operator, but this can lead to the loss of valuable information carried by edge directionality. In this paper, we introduce a new clustering framework, the Parametrized Random Walk Diffusion Kernel Clustering (P-RWDKC), which is suitable for handling both directed and undirected graphs. Our framework is based on the diffusion geometry and the generalized spectral clustering framework. Accordingly, we propose an algorithm that automatically reveals the cluster structure at a given scale, by considering the random walk dynamics associated with a parametrized kernel operator, and by estimating its critical diffusion time. Experiments on $K$-NN graphs constructed from real-world datasets and real-world graphs show that our clustering approach performs well in all tested cases, and outperforms existing approaches in most of them. | A novel clustering algorithm for directed graphs based on the diffusion geometry framework and parametrized random walk operators |
Triplet-based loss functions have been the paradigm of choice for robust deep metric learning (DML). However, conventional triplet-based losses require carefully tuning a decision boundary, i.e., violation margin. When performing online triplet mining on each mini-batch, choosing a good global and constant prior value for violation margin is challenging and irrational. To circumvent this issue, we propose a novel yet efficient concordance-induced triplet (CIT) loss as an objective function to train DML models. We formulate the similarity of triplet samples as a concordance constraint problem, then directly optimize concordance during DML model learning. Triplet concordance refers to the predicted ordering of intra-class and inter-class similarities being correct, which is invariant to any monotone transformation of the decision boundary of triplet samples. Hence, our CIT loss is free from the plague of adopting the violation margin as a prior constraint. In addition, due to the high training complexity of triplet-based losses, we introduce a partial likelihood term for CIT loss to impose additional penalties on hard triplet samples, thus enforcing fast convergence. We extensively experiment on a variety of DML tasks to demonstrate the elegance and simplicity of our CIT loss against its counterparts. In particular, on face recognition, person re-identification, as well as image retrieval datasets, our method can achieve comparable performances with state-of-the-arts without tuning any hyper-parameters laboriously. | A simple and elegant loss function is proposed to exploit the concordance constraint of triplet similarity for deep metric learning. |
Transformers in their common form are inherently limited to operate on whole token sequences rather than on one token at a time. Consequently, their use during online inference on time-series data entails considerable redundancy due to the overlap in successive token sequences. In this work, we propose novel formulations of the Scaled Dot-Product Attention, which enable Transformers to perform efficient online token-by-token inference on a continual input stream. Importantly, our modifications are purely to the order of computations, while the outputs and learned weights are identical to those of the original Transformer Encoder. We validate our Continual Transformer Encoder with experiments on the THUMOS14, TVSeries and GTZAN datasets with remarkable results: Our Continual one- and two-block architectures reduce the floating point operations per prediction by up to 63x and 2.6x, respectively, while retaining predictive performance. | A Transformer Decorder acceleration for online stream processing validated with experiments in Online Action Detection and Audio Classification. |
In graph neural networks (GNNs), both node features and labels are examples of graph signals, a key notion in graph signal processing (GSP). While it is common in GSP to impose signal smoothness constraints in learning and estimation tasks, it is unclear how this can be done for discrete node labels. We bridge this gap by introducing the concept of distributional graph signals. In our framework, we work with the distributions of node labels instead of their values and propose notions of smoothness and non-uniformity of such distributional graph signals. We then propose a general regularization method for GNNs that allows us to encode distributional smoothness and non-uniformity of the model output in semi-supervised node classification tasks. Numerical experiments demonstrate that our method can significantly improve the performance of most base GNN models in different problem settings. | In this paper, we introduce the new notion of distributional graph signals and use it to design a GNN regularization scheme. |
Training machine learning models robust to distribution shifts is critical for real-world applications. Some robust training algorithms (e.g., Group DRO) specialize to group shifts and require group information on all training points. Other methods (e.g., CVaR DRO) that do not need group annotations can be overly conservative, since they naively upweight high loss points which may form a contrived set that does not correspond to any meaningful group in the real world (e.g., when the high loss points are randomly mislabeled training points). In this work, we address limitations in prior approaches by assuming a more nuanced form of group shift: conditioned on the label, we assume that the true group function (indicator over group) is simple. For example, we may expect that group shifts occur along low bitrate features (e.g., image background, lighting). Thus, we aim to learn a model that maintains high accuracy on simple group functions realized by these low bitrate features, that need not spend valuable model capacity achieving high accuracy on contrived groups of examples. Based on this, we consider the two-player game formulation of DRO where the adversary's capacity is bitrate-constrained. Our resulting practical algorithm, Bitrate-Constrained DRO (\bdro), does not require group information on training samples yet matches the performance of Group DRO on datasets that have training group annotations and that of CVaR DRO on long-tailed distributions. Our theoretical analysis reveals that in some settings \bdro objective can provably yield statistically efficient and less conservative solutions than unconstrained CVaR DRO. | Robustness to group shifts without training group annotations can be achieved with a constrained form of DRO. |
Recently, it has been demonstrated that a wide variety of (non) linear two-layer neural networks (such as two-layer perceptrons, convolutional networks, and self-attention) can be posed as equivalent convex optimization problems, with an induced regularizer which encourages low rank. However, this regularizer becomes prohibitively expensive to compute at moderate scales, impeding training convex neural networks. To this end, we propose applying the Burer-Monteiro factorization to convex neural networks, which for the first time enables a Burer-Monteiro perspective on neural networks with non-linearities. This factorization leads to an equivalent yet computationally tractable non-convex alternative with no spurious local minima. We develop a novel relative optimality bound of stationary points of the Burer-Monteiro factorization, thereby providing verifiable conditions under which any stationary point is a global optimum. Further, for the first time, we show that linear self-attention with sufficiently many heads has no spurious local minima. Our experiments demonstrate the utility and implications of the novel relative optimality bound for stationary points of the Burer-Monteiro factorization. | We apply the Burer-Monteiro factorization to two-layer ReLU (fully-connected, convolutional, self-attention) neural networks by leveraging their implicit convexity, and provide insights into stationary points and local optima of these networks. |
In this paper, we propose a new self-supervised method, which is called denoising masked autoencoders (DMAE), for learning certified robust classifiers of images. In DMAE, we corrupt each image by adding Gaussian noises to each pixel value and randomly masking several patches. A Transformer-based encoder-decoder model is then trained to reconstruct the original image from the corrupted one. In this learning paradigm, the encoder will learn to capture relevant semantics for the downstream tasks, which is also robust to Gaussian additive noises. We show that the pre-trained encoder can naturally be used as the base classifier in Gaussian smoothed models, where we can analytically compute the certified radius for any data point. Although the proposed method is simple, it yields significant performance improvement in downstream classification tasks. We show that the DMAE ViT-Base model, which just uses 1/10 parameters of the model developed in recent work (Carlini et al., 2022), achieves competitive or better certified accuracy in various settings. The DMAE ViT-Large model significantly surpasses all previous results, establishing a new state-of-the-art on ImageNet dataset. We further demonstrate that the pre-trained model has good transferability to the CIFAR-10 dataset, suggesting its wide adaptability. Models and code are available at
https://github.com/quanlin-wu/dmae. | In this paper, we propose a new self-supervised method for learning certified robust classifiers of images. |
The overlap index between two probability distributions has various applications in statistics, machine learning, and other scientific research. However, approximating the overlap index is challenging when the probability distributions are unknown (i.e., distribution-free settings). This paper proposes an easy-to-compute upper bound for the overlap index without requiring any knowledge of the distribution models. We first utilize the bound to find the upper limit for the accuracy of a trained machine learning model when a domain shift exists. We additionally employ this bound to study the distribution membership classification of given samples. Specifically, we build a novel, distribution-free, computation-efficient, memory-efficient one-class classifier by converting the bound into a confidence score function. The proposed classifier does not need to train any parameters and requires only a small number of in-class samples to be accurate. The classifier shows its efficacy on various datasets and outperforms many state-of-the-art methods in different one-class classification scenarios, including novelty detection, out-of-distribution detection, and backdoor detection. The obtained results show significant promise toward broadening the applications of overlap-based metrics. | This paper proposes an easy-to-compute upper bound for the overlap index and applies it for domain shift analysis and one-class classification. |
We study the implicit regularization of gradient descent towards structured sparsity via a novel neural reparameterization, which we call a diagonally grouped linear neural network. We show the following intriguing property of our reparameterization: gradient descent over the squared regression loss, without any explicit regularization, biases towards solutions with a group sparsity structure. In contrast to many existing works in understanding implicit regularization, we prove that our training trajectory cannot be simulated by mirror descent. We analyze the gradient dynamics of the corresponding regression problem in the general noise setting and obtain minimax-optimal error rates. Compared to existing bounds for implicit sparse regularization using diagonal linear networks, our analysis with the new reparameterization shows improved sample complexity. In the degenerate case of size-one groups, our approach gives rise to a new algorithm for sparse linear regression. Finally, we demonstrate the efficacy of our approach with several numerical experiments. | We study the implicit regularization of gradient descent towards structured sparsity via a novel neural reparameterization. |
Neural surface reconstruction aims to reconstruct accurate 3D surfaces based on multi-view images. Previous methods based on neural volume rendering mostly train a fully implicit model with MLPs, which typically require hours of training for a single scene. Recent efforts explore the explicit volumetric representation to accelerate the optimization via memorizing significant information with learnable voxel grids. However, existing voxel-based methods often struggle in reconstructing fine-grained geometry, even when combined with an SDF-based volume rendering scheme. We reveal that this is because 1) the voxel grids tend to break the color-geometry dependency that facilitates fine-geometry learning, and 2) the under-constrained voxel grids lack spatial coherence and are vulnerable to local minima. In this work, we present Voxurf, a voxel-based surface reconstruction approach that is both efficient and accurate. Voxurf addresses the aforementioned issues via several key designs, including 1) a two-stage training procedure that attains a coherent coarse shape and recovers fine details successively, 2) a dual color network that maintains color-geometry dependency, and 3) a hierarchical geometry feature to encourage information propagation across voxels. Extensive experiments show that Voxurf achieves high efficiency and high quality at the same time. On the DTU benchmark, Voxurf achieves higher reconstruction quality with a 20x training speedup compared to previous fully implicit methods. Our code is publicly available at https://github.com/wutong16/Voxurf/. | We present Voxurf, a voxel-based approach for efficient and accurate neural surface reconstruction. |
A common approach to transfer learning under distribution shift is to fine-tune the last few layers of a pre-trained model, preserving learned features while also adapting to the new task. This paper shows that in such settings, selectively fine-tuning a subset of layers (which we term surgical fine-tuning) matches or outperforms commonly used fine-tuning approaches. Moreover, the type of distribution shift influences which subset is more effective to tune: for example, for image corruptions, fine-tuning only the first few layers works best. We validate our findings systematically across seven real-world data tasks spanning three types of distribution shifts. Theoretically, we prove that for two-layer neural networks in an idealized setting, first-layer tuning can outperform fine-tuning all layers. Intuitively, fine-tuning more parameters on a small target dataset can cause information learned during pre-training to be forgotten, and the relevant information depends on the type of shift. | Selectively fine-tuning a subset of layers outperforms full fine-tuning when transferring to tasks with various distribution shifts. |
Comprehensive studies of synthetic optical flow datasets have attempted to reveal what properties lead to accuracy improvement in learning-based optical flow estimation. However, manually identifying and verifying the properties that contribute to accurate optical flow estimation require large-scale trial-and-error experiments with iteratively generating whole synthetic datasets and training on them, \ie, impractical. To address this challenge, we propose a differentiable optical flow data generation pipeline and a loss function to drive the pipeline, called DFlow. DFlow efficiently synthesizes a dataset effective for a target domain without the need for cumbersome try-and-errors. This favorable property is achieved by proposing an efficient dataset comparison method that uses neural networks to approximately encode each dataset and compares the proxy networks instead of explicitly comparing datasets in a pairwise way. Our experiments show the competitive performance of our DFlow against the prior arts in pre-training. Furthermore, compared to competing datasets, DFlow achieves the best fine-tuning performance on the Sintel public benchmark with RAFT. | Differentiable and efficient optical flow data generation pipeline |
Improvements in the performance of computing systems, driven by Moore’s Law, have transformed society. As such hardware-driven gains slow down, it becomes even more important for software developers to focus on performance and efficiency during development. While several studies have demonstrated the potential from such improved code efficiency (e.g., 2x better generational improvements compared to hardware), unlocking these gains in practice has been challenging. Reasoning about algorithmic complexity and the interaction of coding patterns on hardware can be challenging for the average programmer, especially when combined with pragmatic constraints around development velocity and multi-person development.
This paper seeks to address this problem. We analyze a large competitive programming dataset from the Google Code Jam competition and find that efficient code is indeed rare, with a 2x runtime difference between the median and the 90th percentile of solutions. We propose using machine learning to automatically provide prescriptive feedback in the form of hints, to guide programmers towards writing high-performance code. To automatically learn these hints from the dataset, we propose a novel discrete variational auto-encoder, where each discrete latent variable represents a different learned category of code-edit that increases performance. We show that this method represents the multi-modal space of code efficiency edits better than a sequence-to-sequence baseline and generates a distribution of more efficient solutions. | We propose a generative model trained on programming competition submissions to transform programs into faster versions of those programs. |
Neighbor embedding methods $t$-SNE and UMAP are the de facto standard for visualizing high-dimensional datasets. Motivated from entirely different viewpoints, their loss functions appear to be unrelated. In practice, they yield strongly differing embeddings and can suggest conflicting interpretations of the same data. The fundamental reasons for this and, more generally, the exact relationship between $t$-SNE and UMAP have remained unclear. In this work, we uncover their conceptual connection via a new insight into contrastive learning methods. Noise-contrastive estimation can be used to optimize $t$-SNE, while UMAP relies on negative sampling, another contrastive method. We find the precise relationship between these two contrastive methods, and provide a mathematical characterization of the distortion introduced by negative sampling. Visually, this distortion results in UMAP generating more compact embeddings with tighter clusters compared to $t$-SNE. We exploit this new conceptual connection to propose and implement a generalization of negative sampling, allowing us to interpolate between (and even extrapolate beyond) $t$-SNE and UMAP and their respective embeddings. Moving along this spectrum of embeddings leads to a trade-off between discrete / local and continuous / global structures, mitigating the risk of over-interpreting ostensible features of any single embedding. We provide a PyTorch implementation. | We show that UMAP is effectively negative sampling applied to the t-SNE loss function. |
We present a general framework for evaluating image counterfactuals. The power and flexibility of deep generative models make them valuable tools for learning mechanisms in structural causal models. However, their flexibility makes counterfactual identifiability impossible in the general case.
Motivated by these issues, we revisit Pearl's axiomatic definition of counterfactuals to determine the necessary constraints of any counterfactual inference model: composition, reversibility, and effectiveness. We frame counterfactuals as functions of an input variable, its parents, and counterfactual parents and use the axiomatic constraints to restrict the set of functions that could represent the counterfactual, thus deriving distance metrics between the approximate and ideal functions. We demonstrate how these metrics can be used to compare and choose between different approximate counterfactual inference models and to provide insight into a model's shortcomings and trade-offs. | We use the axiomatic definition of counterfactual to derive metrics that enable quantifying the correctness of approximate counterfactual inference models. |
Recent breakthroughs based on big/foundation models reveal a vague avenue for AI, that is, \emph{big data, big/foundation models, big learning, $\cdots$}. Following that avenue, here we elaborate on our newly introduced big learning. Specifically, big learning exhaustively exploits the information/tasks inherent in its large-scale \emph{complete/incomplete} training data, by learning to simultaneously model many/all joint/conditional/marginal data distributions (thus named big learning) with one universal foundation model. We reveal that big learning is what existing foundation models are implicitly doing; accordingly, our big learning provides high-level guidance for flexible design and improvements of foundation models. Besides, big learning ($i$) is equipped with great flexibilities for complete/incomplete training data and for customizing trustworthy data tasks; ($ii$) potentially delivers all joint/conditional/marginal data capabilities after training; ($iii$) significantly reduces the training-test gap with improved model generalization; and ($iv$) potentially unifies conventional machine learning paradigms and enables their flexible cooperations, manifested as a universal learning paradigm. Preliminary experiments verified the effectiveness of the presented big learning.
| We propose a new machine learning framework named Big Learning, which exhaustively exploits data information and underlies existing foundation models. |
Changing how pre-trained models behave---e.g., improving their performance on a downstream task or mitigating biases learned during pre-training---is a common practice when developing machine learning systems. In this work, we propose a new paradigm for steering the behavior of neural networks, centered around task vectors. A task vector specifies a direction in the weight space of a pre-trained model, such that movement in that direction improves performance on the task. We build task vectors by subtracting the weights of a pre-trained model from the weights of the same model after fine-tuning on a task. We show that these task vectors can be modified and combined together through arithmetic operations such as negation and addition, and the behavior of the resulting model is steered accordingly. Moreover, task vectors can be added together to improve performance on multiple tasks at once. Finally, when tasks are linked by an analogy relationship of the form ``A is to B as C is to D", combining task vectors from three of the tasks can improve performance on the fourth, even when no data from the fourth task is used for training. | We study a new paradigm for editing pre-trained models, where weight vectors obtained via fine-tuning can be combined to efficiently and effectively steer model behavior. |
Learning models of the environment from pure interaction is often considered an essential component of building lifelong reinforcement learning agents. However, the common practice in model-based reinforcement learning is to learn models that model every aspect of the agent's environment, regardless of whether they are important in coming up with optimal decisions or not. In this paper, we argue that such models are not particularly well-suited for performing scalable and robust planning in lifelong reinforcement learning scenarios and we propose new kinds of models that only model the relevant aspects of the environment, which we call minimal value-equivalent partial models. After providing the formal definitions of these models, we provide theoretical results demonstrating the scalability advantages of performing planning with minimal value-equivalent partial models and then perform experiments to empirically illustrate our theoretical results. Finally, we provide some useful heuristics on how to learn such models with deep learning architectures and empirically demonstrate that models learned in such a way can allow for performing planning that is robust to distribution shifts and compounding model errors. Overall, both our theoretical and empirical results suggest that minimal value-equivalent partial models can provide significant benefits to performing scalable and robust planning in lifelong reinforcement learning scenarios. | We propose new kinds of models to perform scalable and robust planning in lifelong reinforcement learning. |
We tackle the problem of aligning pre-trained large language models (LMs) with human preferences. If we view text generation as a sequential decision-making problem, reinforcement learning (RL) appears to be a natural conceptual framework. However, using RL for LM-based generation faces empirical challenges, including training instability due to the combinatorial action space, as well as a lack of open-source libraries and benchmarks customized for LM alignment. Thus, a question rises in the research community: is RL a practical paradigm for NLP?
To help answer this, we first introduce an open-source modular library, $RL4LMs$ (Reinforcement Learning for Language Models), for optimizing language generators with RL. The library consists of on-policy RL algorithms that can be used to train any encoder or encoder-decoder LM in the HuggingFace library (Wolf et al. 2020) with an arbitrary reward function. Next, we present the $GRUE$ (General Reinforced-language Understanding Evaluation) benchmark, a set of 6 language generation tasks which are supervised not by target strings, but by reward functions which capture automated measures of human preference.GRUE is the first leaderboard-style evaluation of RL algorithms for NLP tasks. Finally, we introduce an easy-to-use, performant RL algorithm, $NLPO$ (Natural Language Policy Optimization)} that learns to effectively reduce the combinatorial action space in language generation. We show 1) that RL techniques are generally better than supervised methods at aligning LMs to human preferences; and 2) that NLPO exhibits greater stability and performance than previous policy gradient methods (e.g., PPO (Schulman et al. 2017)), based on both automatic and human evaluations. | We provide an open-source framework, benchmark, and novel algorithm to train large language models to better align to automated measures of human preferences. |
Recent years have seen significant progress in techniques for learning high-dimensional distributions. Many modern methods, from diffusion models to Energy-Based-Models (EBMs), adopt a coarse-to-fine approach. This is often done by introducing a series of auxiliary distributions that gradually change from the data distribution to some simple distribution (e.g. white Gaussian noise). Methods in this category separately learn each auxiliary distribution (or transition between pairs of consecutive distributions) and then use the learned models sequentially to generate samples. In this paper, we offer a simple way to generalize this idea by treating the ``time'' index of the series as a random variable and framing the problem as that of learning a single joint distribution of "time" and samples. We show that this joint distribution can be learned using any existing EBM method and that it allows achieving improved results. As an example, we demonstrate this approach using contrastive divergence (CD) in its most basic form. On CIFAR-10 and CelebA ($32\times 32$), this method outperforms previous CD-based methods in terms of inception and FID scores. | We generalize the common practice of utilizing a series of auxiliary distributions in EBM training, and utilize this approach to improve the performance of existing methods. |
Recent advancements in explainable machine learning provide effective and faithful solutions for interpreting model behaviors. However, many explanation methods encounter efficiency issues, which largely limit their deployments in practical scenarios. Real-time explainer (RTX) frameworks have thus been proposed to accelerate the model explanation process by learning an one-feed-forward explainer. Existing RTX frameworks typically build the explainer under the supervised learning paradigm, which requires large amounts of explanation labels as the ground truth. Considering that accurate explanation labels are usually hard to obtain, due to constrained computational resources and limited human efforts, effective explainer training is still challenging in practice. In this work, we propose a COntrastive Real-Time eXplanation (CoRTX) framework to learn the explanation-oriented representation and relieve the intensive dependence of explainer training on explanation labels. Specifically, we design a synthetic strategy to select positive and negative instances for explanation representation learning. Theoretical analysis show that our selection strategy can benefit the contrastive learning process on explanation tasks. Experimental results on three real-world datasets further demonstrate the efficiency and efficacy of our proposed CoRTX framework. | Learning real-time model explainer with limited explanation labels. |
Knowledge reuse is essential for continual learning, and current methods attempt to realize it through regularization or experience replay. These two strategies have complementary strengths, e.g., regularization methods are compact, but replay methods can mimic batch training more accurately. At present, little has been done to find principled ways to combine the two methods and current heuristics can give suboptimal performance. Here, we provide a principled approach to combine and improve them by using a recently proposed principle of adaptation, where the goal is to reconstruct the “gradients of the past”, i.e., to mimic batch training by estimating gradients from past data. Using this principle, we design a prior that provably gives better gradient reconstructions by utilizing two types of replay and a quadratic weight-regularizer. This improves performance on standard benchmarks such as Split CIFAR, Split TinyImageNet, and ImageNet-1000. Our work shows that a good combination of replay and regularizer-based methods can be very effective in reducing forgetting, and can sometimes even completely eliminate it. | We propose a new, principled yet practical continual learning method that combines the complementary benefits of function-regularisation, weight-regularisation and experience replay. |
Graph Neural Networks (GNNs) have shown great promise in various graph learning tasks. However, the computational overheads of fitting GNNs to large-scale graphs grow rapidly, posing obstacles to GNNs from scaling up to real-world applications. To tackle this issue, Graph Lottery Ticket (GLT) hypothesis articulates that there always exists a sparse subnetwork/subgraph with admirable performance in GNNs with random initialization. Such a pair of core subgraph and sparse subnetwork (called graph lottery tickets) can be uncovered by iteratively applying a novel sparsification method. While GLT provides new insights for GNN compression, it requires a full pretraining process to obtain graph lottery tickets, which is not universal and friendly to real-world applications. Moreover, the graph sparsification in GLT utilizes sampling techniques, which may result in massive information loss and aggregation failure. In this paper, we explore the searching of graph lottery tickets from a complementary perspective -- transforming a random ticket into a graph lottery ticket, which allows us to more comprehensively explore the relationships between the original network/graph and their sparse counterpart. To achieve this, we propose regularization-based network pruning and hierarchical graph sparsification, leading to our Dual Graph Lottery Ticket (DGLT) framework for a joint sparsification of network and graph. Compared to GLT, our DGLT helps achieve a triple-win situation of graph lottery tickets with high sparsity, admirable performance, and good explainability. More importantly, we rigorously prove that our model can eliminate noise and maintain reliable information in substructures using the graph information bottleneck theory. Extensive experimental results on various graph-related tasks validate the effectiveness of our framework. | This paper generalizes Dual Lottery Ticket Hypothesis (DLTH) to the graph to address information loss and aggregation failure issues caused by sampling-based GNN pruning algorithms |
Traditional recurrent neural networks (RNNs) have a fixed, finite number of memory cells. In theory (assuming bounded range and precision), this limits their formal language recognition power to regular languages, and in practice, RNNs have been shown to be unable to learn many context-free languages (CFLs). In order to expand the class of languages RNNs recognize, prior work has augmented RNNs with a nondeterministic stack data structure, putting them on par with pushdown automata and increasing their language recognition power to CFLs. Nondeterminism is needed for recognizing all CFLs (not just deterministic CFLs), but in this paper, we show that nondeterminism and the neural controller interact to produce two more unexpected abilities. First, the nondeterministic stack RNN can recognize not only CFLs, but also many non-context-free languages. Second, it can recognize languages with much larger alphabet sizes than one might expect given the size of its stack alphabet. Finally, to increase the information capacity in the stack and allow it to solve more complicated tasks with large alphabet sizes, we propose a new version of the nondeterministic stack that simulates stacks of vectors rather than discrete symbols. We demonstrate perplexity improvements with this new model on the Penn Treebank language modeling benchmark. | We show that nondeterministic stack RNNs can learn non-CFLs and languages with surprisingly large alphabets, and we propose a new version that models a stack of vector embeddings. |
Switching costs, which capture the costs for changing policies, are regarded as a critical metric in reinforcement learning (RL), in addition to the standard metric of losses (or rewards). However, existing studies on switching costs (with a coefficient that is strictly positive and is independent of the time horizon) have mainly focused on static RL, where the loss distribution is assumed to be fixed during the learning process, and thus practical scenarios where the loss distribution could be non-stationary or even adversarial are not considered. While adversarial RL better models this type of practical scenarios, an open problem remains: how to develop a provably efficient algorithm for adversarial RL with switching costs? This paper makes the first effort towards solving this problem. First, we provide a regret lower-bound that shows that the regret of any algorithm must be larger than $\tilde{\Omega}( ( H S A )^{1/3} T^{2/3} )$, where $T$, $S$, $A$ and $H$ are the number of episodes, states, actions and layers in each episode, respectively. Our lower bound indicates that, due to the fundamental challenge of switching costs in adversarial RL, the best achieved regret (whose dependency on $T$ is $\tilde{O}(\sqrt{T})$) in static RL with switching costs (as well as adversarial RL without switching costs) is no longer achievable. Moreover, we propose two novel switching-reduced algorithms with regrets that match our lower bound when the transition function is known, and match our lower bound within a small factor of $\tilde{O}( H^{1/3} )$ when the transition function is unknown. Our regret analysis demonstrates the near-optimal performance of them. | This paper provides the first algorithms with near-optimal regrets for adversarial reinforcement learning with switching costs, and a matching lower bound on the regret. |
Implicit neural networks have become increasingly attractive in the machine learning community since they can achieve competitive performance but use much less computational resources. Recently, a line of theoretical works established the global convergences for first-order methods such as gradient descent if the implicit networks are over-parameterized. However, as they train all layers together, their analyses are equivalent to only studying the evolution of the output layer. It is unclear how the implicit layer contributes to the training. Thus, in this paper, we restrict ourselves to only training the implicit layer. We show that global convergence is guaranteed, even if only the implicit layer is trained. On the other hand, the theoretical understanding of when and how the training performance of an implicit neural network can be generalized to unseen data is still under-explored. Although this problem has been studied in standard feed-forward networks, the case of implicit neural networks is still intriguing since implicit networks theoretically have infinitely many layers. Therefore, this paper investigates the generalization error for implicit neural networks. Specifically, we study the generalization of an implicit network activated by the ReLU function over random initialization. We provide a generalization bound that is initialization sensitive. As a result, we show that gradient flow with proper random initialization can train a sufficient over-parameterized implicit network to achieve arbitrarily small generalization errors. | This paper analyzes the training and generalization for a implicit neural network with random initialization. |
Current state-of-the-art deep networks are all powered by backpropagation. In this paper, we explore alternatives to full backpropagation in the form of blockwise learning rules, leveraging the latest developments in self-supervised learning. Notably, we show that a blockwise pretraining procedure consisting of training independently the 4 main blocks of layers of a ResNet-50 with Barlow Twins loss function at each block performs almost as well as end-to-end backpropagation on ImageNet: a linear probe trained on top of our blockwise pretrained model obtains a top-1 classification accuracy of 70.48\%, only 1.1\% below the accuracy of an end-to-end pretrained network (71.57\% accuracy). We perform extensive experiments to understand the impact of different components within our method and explore a variety of adaptations of self-supervised learning to the blockwise paradigm, building an exhaustive understanding of the critical avenues for scaling local learning rules to large networks, with implications ranging from hardware design to neuroscience. | We extend current self-supervised learning methods to blockwise training scheme |
We study approximation algorithms for the socially fair $(\ell_p, k)$-clustering problem with $m$ groups which include the socially fair $k$-median ($p=1$) and $k$-means ($p=2$). We present (1) a polynomial-time $(5+2\sqrt{6})^p$-approximation with at most $k+m$ centers (2) a $(5+2\sqrt{6}+\epsilon)^p$-approximation with $k$ centers in time $(nk)^{{2^{O(p)} m^2}/\epsilon}$, and (3) a $(15+6\sqrt{6})^p$ approximation with $k$ centers in time $k^{m}\cdot\text{poly}(n)$. The former is obtained by a refinement of the iterative rounding method via a sequence of linear programs. The latter two are obtained by converting a solution with up to $k+m$ centers to one with $k$ centers by sparsification methods for (2) and via an exhaustive search for (3). We also compare the performance of our algorithms with existing approximation algorithms on benchmark datasets, and find that our algorithms outperform existing methods. | We present a constant-factor approximation algorithm for the socially fair k-means and k-medians problems. |
We introduce adaptive weight decay, which automatically tunes the hyper-parameter for weight decay during each training iteration. For classification problems, we propose changing the value of the weight-decay hyper-parameter on the fly based on the strength of updates from the classification loss (i.e., gradient of cross-entropy), and the regularization loss (i.e., $\ell_2$-norm of the weights). We show that this simple modification can result in large improvements in adversarial robustness — an area which suffers from robust overfitting — without requiring extra data. Specifically, our reformulation results in 20% relative robustness improvement for CIFAR-100, and 10% relative robustness improvement on CIFAR-10 comparing to traditional weight-decay. In addition, this method has other desirable properties, such as less sensitivity to learning rate, and smaller weight norms, which the latter contributes to robustness to overfitting to label noise, and pruning. | We tune the hyper-parameter for weight decay during each iteration to stabilize training networks with smaller weight-norms which results in more robustness to adversarial examples and label noise, and less sensitivity to choices of learning rate. |
Generative models have grown into the workhorse of many state-of-the-art machine learning methods. However, their vulnerability under poisoning attacks has been largely understudied. In this work, we investigate this issue in the context of continual learning, where generative replayers are utilized to tackle catastrophic forgetting. By developing a novel customization of dirty-label input-aware backdoor to the online setting, our attacker manages to stealthily promote forgetting while retaining high accuracy at the current task and sustaining strong defenders. Our approach taps into an intriguing property of generative models, namely that they cannot well capture input-dependent triggers. Experiments on four standard datasets corroborate the poisoner’s effectiveness.
| We develop a novel poisoning attack on generative models to promotes catastrophic forgetting. |
The demand for memory to store model-related statistics ("model memory") is a major bottleneck for training large neural networks. A promising solution is low-precision optimizers, which reduce the numerical precision of the model memory. However, existing work only compresses the momentum, resulting in suboptimal memory efficiency. This paper proposes Low-Precision Model Memory (LPMM), an optimization framework with the entire model memory kept in low precision. LPMM compresses not only the momentum but also model parameters and gradient accumulators. We identify arithmetic underflow as the main problem in building low-precision optimizers and propose a stochastic quantization method and a microbatching technique to overcome this problem. We analyze the convergence behavior of LPMM and theoretically show how the proposed techniques could affect underflowing, which in turn affects the convergence. We apply LPMM to the SGD optimizer with momentum (SGDM). On several realistic benchmarks, LPMM-SGDM can train neural networks with negligible loss of accuracy while reducing over 70% of the model memory compared to the full-precision SGDM. | We propose memory efficient optimizers for deep learning which keep model parameters, momentum and gradient accumulators in low numerical precision. |
Recently, large-scale text retrieval has made impressive progress, facilitating both information retrieval and downstream knowledge-intensive tasks (e.g., open-domain QA and dialogue). With a moderate amount of data, a neural text retriever can outperform traditional methods such as BM25 by a large step. However, while being applied to out-of-domain data, the performance of a neural retriever degrades considerably. Therefore, how to enable a retriever to perform more robustly across different domains or tasks and even show strong zero-shot transfer ability is critical for building scalable IR systems. To this end, we propose HypeR, a hyper-prompted training mechanism to enable uniform retrieval across tasks of different domains. Specifically, our approach jointly trains the query encoder with a shared prompt-based parameter pool and a prompt synthesizer that dynamically composes hyper-prompt for encoding each query from different tasks or domains. Besides, to avoid the mode collapse of prompt attention distribution for different queries, we design a contrastive prompt regularization that promotes the mode of prompt attention to be aligned and uniform. Through multi-task hyper-prompted training, our retriever can master the ability to dynamically represent different types of queries and transfer knowledge across different domains and tasks. Extensive experiments show our model attains better retrieval performance across different tasks and better zero-shot transfer ability compared with various previous methods. | A multitask hyper-prompted training mechanism that enables a neural retriever to dynamically process different types of queries with different hyper-prompts and transfer learned knowledge across different domains and tasks. |
Endowing deep learning models with symmetry priors can lead to a considerable performance improvement. As an interesting bridge between physics and deep learning, the equivariant partial differential operators (PDOs) have drawn much researchers' attention recently. However, to ensure the PDOs translation equivariance, previous works have to require coefficient matrices to be constant and spatially shared for their linearity, which could lead to the sub-optimal feature learning at each position. In this work, we propose a novel nonlinear PDOs scheme that is both spatially adaptive and translation equivariant. The coefficient matrices are obtained by local features through a generator rather than spatially shared. Besides, we establish a new theory on incorporating more equivariance like rotations for such PDOs. Based on our theoretical results, we efficiently implement the generator with an equivariant multilayer perceptron (EMLP). As such equivariant PDOs are generated by neural networks, we call them Neural ePDOs. In experiments, we show that our method can significantly improve previous works with smaller model size in various datasets. Especially, we achieve the state-of-the-art performance on the MNIST-rot dataset with only half parameters of the previous best model. | We propose a novel spatial adaptive equivariant PDOs-based network which achieves superior performance than previous works. |
Recently, Mixture-of-Experts (MoE) has demonstrated success in scaling models to have large amounts of parameters without significant increases in computational cost. However, MoEs have been also reported to be parameter inefficient such that larger models do not always lead to better performance.
In this work, we study how to build parameter-efficient MoE models. Our analysis identifies that MoE layers exhibit poor gradient flow as the number of experts increases, leading to insufficient training of experts. To overcome this issue, we propose a new MoE architecture design (SaMoE), which improves the parameter efficiency of MoE models by learning a soft combination of a global set of expert layers for each MoE layer. Such a scheme enables substantial parameter savings on MoE while achieving comparable or better accuracy than the standard MoE training baseline. Extensive experiments on billion-scale GPT-3 style autoregressive MoE language models demonstrate that SaMoE significantly improves the parameter efficiency of MoE models by reducing up to 5.2X total parameters while obtaining superior pre-training and zero-shot generalization results as compared to baseline. | SaMoE is a parameter efficient MoE architecture design that enables parameter savings on MoE while achieving comparable or better accuracy. |
One of the most common ways children learn when unfamiliar with the environment is by mimicking adults. Imitation learning concerns an imitator learning to behave in an unknown environment from an expert's demonstration; reward signals remain latent to the imitator. This paper studies imitation learning through causal lenses and extends the analysis and tools developed for behavior cloning (Zhang, Kumor, Bareinboim, 2020) to inverse reinforcement learning. First, we propose novel graphical conditions that allow the imitator to learn a policy performing as well as the expert's behavior policy, even when the imitator and the expert's state-action space disagree, and unobserved confounders (UCs) are present. When provided with parametric knowledge about the unknown reward function, such a policy may outperform the expert's. Also, our method is easily extensible and allows one to leverage existing IRL algorithms even when UCs are present, including the multiplicative-weights algorithm (MWAL) (Syed & Schapire, 2008) and the generative adversarial imitation learning (GAIL) (Ho & Ermon, 2016). Finally, we validate our framework by simulations using real-world and synthetic data. | This paper proposes novel inverse reinforcement learning methods to learn effective imitating policies from the expert's demonstrations when unobserved confounders are present. |
Structural information of phylogenetic tree topologies plays an important role in phylogenetic inference. However, finding appropriate topological structures for specific phylogenetic inference tasks often requires significant design effort and domain expertise. In this paper, we propose a novel structural representation method for phylogenetic inference based on learnable topological features. By combining the raw node features that minimize the Dirichlet energy with modern graph representation learning techniques, our learnable topological features can provide efficient structural information of phylogenetic trees that automatically adapts to different downstream tasks without requiring domain expertise. We demonstrate the effectiveness and efficiency of our method on a simulated data tree probability estimation task and a benchmark of challenging real data variational Bayesian phylogenetic inference problems. | Novel phylogenetic inference methods based on learnable topological features via graph neural networks |
The recent success of Vision Transformers is shaking the long dominance of Convolutional Neural Networks (CNNs) in image recognition for a decade. Specifically, in terms of robustness on out-of-distribution samples, recent research finds that Transformers are inherently more robust than CNNs, regardless of different training setups. Moreover, it is believed that such superiority of Transformers should largely be credited to their \emph{self-attention-like architectures per se}. In this paper, we question that belief by closely examining the design of Transformers. Our findings lead to three highly effective architecture designs for boosting robustness, yet simple enough to be implemented in several lines of code, namely a) patchifying input images, b) enlarging kernel size, and c) reducing activation layers and normalization layers. Bringing these components together, we are able to build pure CNN architectures without any attention-like operations that are as robust as, or even more robust than, Transformers. We hope this work can help the community better understand the design of robust neural architectures. The code is publicly available at https://github.com/UCSC-VLAA/RobustCNN. | we show CNNs can be as robust as, or even more robust than, Transformers |
Recently, SyncMap (2021) pioneered an approach to learn complex structures from sequences as well as adapt to any changes in underlying structures. Such approach, inspired by neuron group behaviors, is achieved by using self-organizing dynamical equations without any loss functions. Here we propose Symmetrical SyncMap that goes beyond the original work to show how to create dynamical equations and attractor-repeller points which are stable over the long run, even dealing with imbalanced continual general chunking problems (CGCPs). The main idea is to apply equal updates from positive and negative feedback loops by symmetrical activation. We then introduce the concept of memory window to allow for more positive updates. Our algorithm surpasses or ties other unsupervised state-of-the-art baselines in all 12 imbalanced CGCPs with various difficulties, including dynamical ones. To verify its performance in real-world scenarios, we conduct experiments on several well-studied structure learning problems. The proposed method surpasses substantially other methods in all scenarios, suggesting that symmetrical activation plays a critical role in uncovering topological structures and even hierarchies encoded in temporal data. | Null |
Large-scale transformer models have become the de-facto architectures for various machine learning applications, e.g., CV and NLP.
However, those large models also introduce prohibitive training costs.
To mitigate this issue, we propose a novel random and layerwise token dropping method (\OURS), which skips the computation of a subset of the input tokens at all middle layers.
Particularly, \OURS achieves considerable speedups and comparable accuracy as the standard training baseline.
Compared to other token dropping methods, \OURS does not require (1) any importance score-based metrics, (2) any special token treatment (e.g., \texttt{[CLS]}), and (3) many layers in full sequence length training except the first and the last layers.
Besides, a new \layertoken learning rate schedule is proposed for pretraining problems that resolve the heavy tuning requirement for our proposed training mechanism.
Finally, we demonstrate that \OURS can be applied to broader applications, including \gpt and \bert pretraining as well as ViT and \gpt finetuning tasks.
Our results show that \OURS can save about 33.3\% theoretical compute cost and 25.6\% wall-clock training time while achieving similar zero-shot evaluations on \gptb as compared to baseline. | We present a novel random and layerwise token dropping method that can save up to 33.3% theoretical compute cost and 25.6% wall-clock time while achieving comparable accuracy as compared to the standard training procedure. |
The performance of deep neural networks is strongly influenced by the training dataset setup. In particular, when attributes with a strong correlation with the target attribute are present, the trained model can provide unintended prejudgments and show significant inference errors (i.e., the dataset bias problem). Various methods have been proposed to mitigate dataset bias, and their emphasis is on weakly correlated samples, called bias-conflicting samples. These methods are based on explicit bias labels provided by humans. However, such methods require human costs. Recently, several studies have sought to reduce human intervention by utilizing the output space values of neural networks, such as feature space, logits, loss, or accuracy. However, these output space values may be insufficient for the model to understand the bias attributes well. In this study, we propose a debiasing algorithm leveraging gradient called Per-sample Gradient-based Debiasing (PGD). PGD is comprised of three steps: (1) training a model on uniform batch sampling, (2) setting the importance of each sample in proportion to the norm of the sample gradient, and (3) training the model using importance-batch sampling, whose probability is obtained in step (2). Compared with existing baselines for various datasets, the proposed method showed state-of-the-art accuracy for the classification task. Furthermore, we describe theoretical understandings of how PGD can mitigate dataset bias.
| We solve the dataset bias problem by using the per-sample gradient. Furthermore, we provide the mathematical background of the proposed algorithm. |
We propose schemas as a model for abstractions that can be used for rapid transfer learning, inference, and planning. Common structured representations of concepts and behaviors---schemas---have been proposed as a powerful way to encode abstractions. Latent graph learning is emerging as a new computational model of the hippocampus to explain map learning and transitive inference. We build on this work to show that learned latent graphs in these models have a slot structure---schemas---that allow for quick knowledge transfer across environments. In a new environment, an agent can rapidly learn new bindings between the sensory stream to multiple latent schemas and select the best fitting one to guide behavior. To evaluate these graph schemas, we use two previously published challenging tasks: the memory \& planning game and one-shot StreetLearn, that are designed to test rapid task solving in novel environments. Graph schemas can be learned in far fewer episodes than previous baselines, and can model and plan in a few steps in novel variations of these tasks. We further demonstrate learning, matching, and reusing graph schemas in navigation tasks in more challenging environments with aliased observations and size variations, and show how different schemas can be composed to model larger environments. | We propose schemas in a higher order graph structures as a model for abstractions that can be used for rapid transfer learning, inference, and planning. |
Diffusion models have achieved unprecedented performance in generative modeling. The commonly-adopted formulation of the latent code of diffusion models is a sequence of gradually denoised samples, as opposed to the simpler (e.g., Gaussian) latent space of GANs, VAEs, and normalizing flows. This paper provides an alternative, Gaussian formulation of the latent space of various diffusion models, as well as an invertible DPM-Encoder that maps images into the latent space. While our formulation is purely based on the definition of diffusion models, we demonstrate several intriguing consequences. (1) Empirically, we observe that a common latent space emerges from two diffusion models trained independently on related domains. In light of this finding, we propose CycleDiffusion, which uses DPM-Encoder for unpaired image-to-image translation. Furthermore, applying CycleDiffusion to text-to-image diffusion models, we show that large-scale text-to-image diffusion models can be used as zero-shot image-to-image editors. (2) One can guide pre-trained diffusion models and GANs by controlling the latent codes in a unified, plug-and-play formulation based on energy-based models. Using the CLIP model and a face recognition model as guidance, we demonstrate that diffusion models have better coverage of low-density sub-populations and individuals than GANs. | By defining diffusion models' Gaussian latent spaces, they can be used like CycleGANs and GANs and beyond (e.g., zero-shot image-to-image translation with text-to-image diffusion models) |
Non-autoregressive translation (NAT) reduces the decoding latency but suffers from performance degradation due to the multi-modality problem. Recently, the structure of directed acyclic graph has achieved great success in NAT, which tackles the multi-modality problem by introducing dependency between vertices. However, training it with negative log-likelihood loss implicitly requires a strict alignment between reference tokens and vertices, weakening its ability to handle multiple translation modalities. In this paper, we hold the view that all paths in the graph are fuzzily aligned with the reference sentence. We do not require the exact alignment but train the model to maximize a fuzzy alignment score between the graph and reference, which takes captured translations in all modalities into account. Extensive experiments on major WMT benchmarks show that our method substantially improves translation performance and increases prediction confidence, setting a new state of the art for NAT on the raw training data. | We introduce a fuzzy alignment objective in Directed Acyclic Graph for NAT, setting a new state of the art for NAT on the raw training data. |
Large Language Models (LLMs) have been the subject of active research, significantly advancing the field of Natural Language Processing (NLP). From BERT to BLOOM, LLMs have surpassed state-of-the-art results in various natural language tasks such as question answering, summarization, and text generation. Many ongoing efforts are focused on understanding LLMs' capabilities, including their knowledge of the world, syntax, and semantics. However, extending the textual prowess of LLMs to symbolic reasoning has been slow and predominantly focused on tackling problems related to the mathematical field. In this paper, we explore the use of LLMs for automated planning - a branch of AI concerned with the realization of action sequences (plans) to achieve a goal, typically for execution by intelligent agents, autonomous robots, and unmanned vehicles. We introduce Plansformer; an LLM fine-tuned on planning problems and capable of generating plans with favorable behavior in terms of correctness and length with minimal knowledge-engineering efforts. We also demonstrate the adaptability of Plansformer in solving different planning domains with varying complexities, owing to the transfer learning abilities of LLMs. For one configuration of Plansformer, we achieve ~97\% valid plans, out of which ~95\% are optimal for Towers of Hanoi - a puzzle-solving domain. | Plansformer, an LLM fine-tuned on planning problems and capable of generating plans with favorable behavior in terms of correctness and length with minimal knowledge-engineering efforts. |
Neural networks can be fragile to input noise and adversarial attacks. In this work, we consider Neural Ordinary Differential Equations (NODEs) – a family of continuous-depth neural networks represented by dynamical systems - and propose to use contraction theory to improve their robustness. A dynamical system is contractive if two trajectories starting from different initial conditions converge to each other exponentially fast. Contractive NODEs can enjoy increased robustness as slight perturbations of the features do not cause a significant change in the output. Contractivity can be induced during training by using a regularization term involving the Jacobian of the system dynamics. To reduce the computational burden, we show that it can also be promoted using carefully selected weight regularization terms for a class of NODEs with slope-restricted activation functions, including convolutional networks commonly used in image classification. The performance of the proposed regularizers is illustrated through benchmark image classification tasks on MNIST and FashionMNIST datasets, where images are corrupted by different kinds of noise and attacks. | We use contraction theory for dynamical systems to design regularizers improving the robustness of neural ODEs. |
Multiple instance learning (MIL) is a popular weakly-supervised learning model on the whole slide image (WSI) for AI-assisted pathology diagnosis. The recent advance in attention-based MIL allows the model to find its region-of-interest (ROI) for interpretation by learning the attention weights for image patches of WSI slides. However, we empirically find that the interpretability of some related methods is either untrustworthy as the principle of MIL is violated or unsatisfactory as the high-attention regions are not consistent with experts' annotations. In this paper, we propose Bayes-MIL to address the problem from a probabilistic perspective. The induced patch-level uncertainty is proposed as a new measure of MIL interpretability, which outperforms previous methods in matching doctors annotations. We design a slide-dependent patch regularizer (SDPR) for the attention, imposing constraints derived from the MIL assumption, on the attention distribution. SDPR explicitly constrains the model to generate correct attention values. The spatial information is further encoded by an approximate convolutional conditional random field (CRF), for better interpretability. Experimental results show Bayes-MIL outperforms the related methods in patch-level and slide-level metrics and provides much better interpretable ROI on several large-scale WSI datasets. | Bayesian modeling of multiple instance learning addresses untrustworthy and unsatisfactory interpretability problem of related methods. |
Hyperbolic graph convolutional networks have been successfully applied to represent complex graph data structures. However, optimization on Riemannian manifolds is nontrivial thus most of the existing hyperbolic networks build the network operations on the tangent space of the manifold, which is a Euclidean local approximation. This distorts the learnt features, limits the representation capacity of the network and makes it hard to optimize the network. In this work, we introduce a fully hyperbolic graph convolutional network (GCN), referred to as SRBGCN, which performs neural computations such as feature transformation and aggregation directly on the manifold, using manifold-preserving Lorentz transformations that include spatial rotation (SR) and boost (B) operations. Experiments conducted on static graph datasets for node classification and link prediction tasks validate the performance of the proposed method. | This work introduces a fully hyperbolic network that uses direct Lorentz transformations to learn the features directly on the manifold. |
The unprecedented growth in DNN model complexity, size and the amount of training data have led to a commensurate increase in demand for computing and a search for minimal encoding. Recent research advocates Hybrid Block Floating-Point (HBFP) as a technique that minimizes silicon provisioning in accelerators by converting the majority of arithmetic operations in training to 8-bit fixed-point. In this paper, we perform a full-scale exploration of the HBFP design space including minimal mantissa encoding, varying block sizes, and mixed mantissa bit-width across layers and epochs. We propose Accuracy Boosters, an epoch-driven mixed-mantissa HBFP that uses 6-bit mantissa only in the last epoch and converts 99.7% of all arithmetic operations in training to 4-bit mantissas. Accuracy Boosters enable reducing silicon provisioning for an HBFP training accelerator by 16.98× as compared to FP32, while preserving or outperforming FP32 accuracy. | We propose an epoch-driven mixed-mantissa Hybrid Block Floating-Point training method converting 99.7% of arithmetic operations in training to 4-bit mantissas and using 6-bit mantissas in the last epoch, while preserving/outperforming FP32 accuracy. |
While instance-level explanation of GNN is a well-studied problem with plenty of approaches being developed, providing a global explanation for the behaviour of a GNN is much less explored, despite its potential in interpretability and debugging. Existing solutions either simply list local explanations for a given class, or generate a synthetic prototypical graph with maximal score for a given class, completely missing any combinatorial aspect that the GNN could have learned.
In this work, we propose GLGExplainer (Global Logic-based GNN Explainer), the first Global Explainer capable of generating explanations as arbitrary Boolean combinations of learned graphical concepts. GLGExplainer is a fully differentiable architecture that takes local explanations as inputs and combines them into a logic formula over graphical concepts, represented as clusters of local explanations.
Contrary to existing solutions, GLGExplainer provides accurate and human-interpretable global explanations that are perfectly aligned with ground-truth explanations (on synthetic data) or match existing domain knowledge (on real-world data). Extracted formulas are faithful to the model predictions, to the point of providing insights into some occasionally incorrect rules learned by the model, making GLGExplainer a promising diagnostic tool for learned GNNs. | We propose GLGExplainer, the first Global Explainer for GNNs capable of generating explanations as arbitrary Boolean combinations of graphical concepts. |
Self-supervised learning has achieved significant success in learning visual representations without the need for manual annotation. To obtain generalizable representations, a meticulously designed data augmentation strategy is one of the most crucial parts. Recently, multi-crop strategies utilizing a set of small crops as positive samples have been shown to learn spatially structured features. However, it overlooks the diverse contextual backgrounds, which reduces the variance of the input views and degenerates the performance. To address this problem, we propose a mosaic representation learning framework (MosRep), consisting of a new data augmentation strategy that enriches the backgrounds of each small crop and improves the quality of visual representations. Specifically, we randomly sample numbers of small crops from different input images and compose them into a mosaic view, which is equivalent to introducing different background information for each small crop. Additionally, we further jitter the mosaic view to prevent memorizing the spatial locations of each crop. Along with optimization, our MosRep gradually extracts more discriminative features. Extensive experimental results demonstrate that our method improves the performance far greater than the multi-crop strategy on a series of downstream tasks, e.g., +7.4% and +4.9% than the multi-crop strategy on ImageNet-1K with 1% label and 10% label, respectively. Code is available at https://github.com/DerrickWang005/MosRep.git.
| We propose a simple and effective mosaic representation learning framework consisting of a new data augmentation strategy, which aims to adequately learn discriminative feature representations. |
In streaming settings, speech recognition models have to map sub-sequences of speech to text before the full audio stream becomes available. However, since alignment information between speech and text is rarely available during training, models need to learn it in a completely self-supervised way. In practice, the exponential number of possible alignments makes this extremely challenging, with models often learning peaky or sub-optimal alignments. Prima facie, the exponential nature of the alignment space makes it difficult to even quantify the uncertainty of a model's alignment distribution. Fortunately, it has been known for decades that the entropy of a probabilistic finite state transducer can be computed in time linear to the size of the transducer via a dynamic programming reduction based on semirings. In this work, we revisit the entropy semiring for neural speech recognition models, and show how alignment entropy can be used to supervise models through regularization or distillation. We also contribute an open-source implementation of CTC and RNN-T in the semiring framework that includes numerically stable and highly parallel variants of the entropy semiring. Empirically, we observe that the addition of alignment distillation improves the accuracy and latency of an already well-optimized teacher-student distillation model, achieving state-of-the-art performance on the Librispeech dataset in the streaming scenario. | A numerically stable open-source implementation of the entropy semiring for CTC and RNN-T; obtained SOTA on Librispeech streaming. |
The necessity for cooperation among intelligent machines has popularised cooperative multi-agent reinforcement learning (MARL) in the artificial intelligence (AI) research community. However, many research endeavours have been focused on developing practical MARL algorithms whose effectiveness has been studied only empirically, thereby lacking theoretical guarantees. As recent studies have revealed, MARL methods often achieve performance that is unstable in terms of reward monotonicity or suboptimal at convergence. To resolve these issues, in this paper, we introduce a novel framework named Heterogeneous-Agent Mirror Learning (HAML) that provides a general template for MARL algorithmic designs. We prove that algorithms derived from the HAML template satisfy the desired properties of the monotonic improvement of the joint reward and the convergence to Nash equilibrium. We verify the practicality of HAML by proving that the current state-of-the-art cooperative MARL algorithms, HATRPO and HAPPO, are in fact HAML instances. Next, as a natural outcome of our theory, we propose HAML extensions of two well-known RL algorithms, HAA2C (for A2C) and HADDPG (for DDPG), and demonstrate their effectiveness against strong baselines on StarCraftII and Multi-Agent MuJoCo tasks. | A general theoretical framework for development of multi-agent reinforcement learning algorithms. |
Monocular Depth Estimation (MDE) is a critical component in applications such as autonomous driving. There are various attacks against MDE networks. These attacks, especially the physical ones, pose a great threat to the security of such systems. Traditional adversarial training method requires ground-truth labels and hence cannot be directly applied to self-supervised MDE that does not have depth ground truth. Some self-supervised model hardening technique (e.g., contrastive learning) ignores the domain knowledge of MDE and can hardly achieve optimal performance. In this work, we propose a novel adversarial training method for self-supervised MDE models based on view synthesis without using the depth ground truth. We improve adversarial robustness against physical-world attacks using $L_0$-norm-bounded perturbation in training. We compare our method with supervised learning-based and contrastive learning-based methods that are tailored for MDE. Results on two representative MDE networks show that we achieve better robustness against various adversarial attacks with nearly no benign performance degradation. | Use self-supervised adversarial training to harden monocular depth estimation models against physical-world adversarial attacks. |
Multi-modal multi-task learning (M$^2$TL) aims to discover the implicit correspondences among heterogeneous modalities and tasks, which is common in real-world applications like autonomous driving and robotics control. Current single-model solutions for M$^2$TL usually fall short in several aspects. The shared backbone between the modalities is prone to overfitting the simpler modality, while jointly optimizing the tasks suffers from unstable training due to the gradient conflicts across tasks. On the other hand, designing a separate model for each task and modality can avoid the above problems but leads to prohibitively expensive computation and memory consumption, rendering this approach unrealistic.
In this work, we propose M$^3$SAT, a sparsely activated transformer for efficient M$^2$TL. The proposed framework tailors the mixture-of-experts (MoEs) into both the self-attention and the feed-forward networks (FFN) of a transformer backbone. It adopts the routing policy to assign attention-heads and FFN experts during training, which effectively disentangles the parameter space to prevent training conflicts among diverse modalities and tasks. Meanwhile, disentangled parameter space also restrains the problem of simple modal prone to overfitting. Sparsely activating the transformer also enables efficient computation for each input sample. Through comprehensive evaluation, we demonstrate the effectiveness of our M$^3$SAT: a remarkable performance margin (\textit{e.g.}, $\ge 1.37\%$) is achieved over the dense models with the same computation cost. More importantly, M$^3$SAT can achieve the above performance improvements with a fraction of the computation cost -- our computation is only $1.38\% \sim 53.51\%$ of that of the SOTA methods. Our code will be released upon acceptance. | Adapt the mixture-of-experts (MoEs) into both the self-attention and the feed-forward networks (FFN) of a transformer backbone for efficient multi-task learning from multiple modalities. |
Limited compute, memory, and communication capabilities of edge users create a significant bottleneck for federated learning (FL) of large models. Current literature typically tackles the challenge with a heterogeneous client setting or allows training to be offloaded to the server. However, the former requires a fraction of clients to train near-full models, which may not be achievable at the edge; while the latter can compromise privacy with sharing of intermediate representations or labels. In this work, we consider a realistic, but much less explored, cross-device FL setting in which no client has the capacity to train a full large model nor is willing to share any intermediate representations with the server. To this end, we present Principal Sub-Model (PriSM) training methodology, which leverages models’ low-rank structure and kernel orthogonality to train sub-models in the orthogonal kernel space. More specifically, by applying singular value decomposition to original kernels in the server model, PriSM first obtains a set of principal orthogonal kernels with importance weighed by their singular values. Thereafter, PriSM utilizes a novel sampling strategy that selects different subsets of the principal kernels independently to create sub-models for clients with reduced computation and communication requirements. Importantly, a kernel with a large singular value is assigned with a high sampling probability. Thus, each sub-model is a low-rank approximation of the full large model, and all clients together achieve nearly full coverage of the principal kernels. To further improve memory efficiency, PriSM exploits low-rank structure in intermediate representations and allows each sub-model to learn only a subset of them while still preserving training performance. Our extensive evaluations on multiple datasets in various resource-constrained settings demonstrate that PriSM can yield an improved performance of up to $10\%$ compared to existing alternatives, when training sub-models with only $20\%$ principal kernels ($\sim 5\%$ of the full server model.). | We provide a sub-model training method that enabled resource-constrained clients to train large model in FL. |
This paper asks the intriguing question: is it possible to exploit neural architecture search (NAS) as a new attack vector to launch previously improbable attacks? Specifically, we present EVAS, a new attack that leverages NAS to find neural architectures with inherent backdoors and exploits such vulnerability using input-aware triggers. Compared with existing attacks, EVAS demonstrates many interesting properties: (i) it does not require polluting training data or perturbing model parameters; (ii) it is agnostic to downstream fine-tuning or even re-training from scratch; (iii) it naturally evades defenses that rely on inspecting model parameters or training data. With extensive evaluation on benchmark datasets, we show that EVAS features high evasiveness, transferability, and robustness, thereby expanding the adversary's design spectrum. We further characterize the mechanisms underlying EVAS, which are possibly explainable by architecture-level ``shortcuts'' that recognize trigger patterns. This work showcases that NAS can be exploited in a harmful way to find architectures with inherent backdoor vulnerability. The code is available at https://github.com/ain-soph/nas_backdoor. | This paper presents EVAS, a new attack to leverage NAS to find neural architectures with exploitable backdoor vulnerability. |
Unlearning has emerged as a technique to efficiently erase information of deleted records from learned models. We show, however, that the influence created by the original presence of a data point in the training set can still be detected after running certified unlearning algorithms (which can result in its reconstruction by an adversary). Thus, under realistic assumptions about the dynamics of model releases over time and in the presence of adaptive adversaries, we show that unlearning is not equivalent to data deletion and does not guarantee the "right to be forgotten." We then propose a more robust data-deletion guarantee and show that it is necessary to satisfy differential privacy to ensure true data deletion. Under our notion, we propose an accurate, computationally efficient, and secure data-deletion machine learning algorithm in the online setting based on noisy gradient descent algorithm. | We show that unlearning guarantees do not ensure the "right to be forgotten" in the online setting, and we certify noisy GD as a trustworthy and utility-preserving deletion algorithm under our improved notion of data deletion. |
We present a novel neural-networks-based algorithm to compute optimal transport maps and plans for strong and weak transport costs. To justify the usage of neural networks, we prove that they are universal approximators of transport plans between probability distributions. We evaluate the performance of our optimal transport algorithm on toy examples and on the unpaired image-to-image translation. | We present a novel neural-networks-based algorithm to compute optimal transport maps and plans for strong and weak transport costs. |
Deep Neural Networks are known to be brittle to even minor distribution shifts compared to the training distribution. While one line of work has demonstrated that \emph{Simplicity Bias} (SB) of DNNs -- bias towards learning only the simplest features -- is a key reason for this brittleness, another recent line of work has surprisingly found that diverse/ complex features are indeed learned by the backbone, and their brittleness is due to the linear classification head relying primarily on the simplest features. To bridge the gap between these two lines of work, we first hypothesize and verify that while SB may not altogether preclude learning complex features, it amplifies simpler features over complex ones. Namely, simple features are replicated several times in the learned representations while complex features might not be replicated. This phenomenon, we term \emph{Feature Replication Hypothesis}, coupled with the \emph{Implicit Bias} of SGD to converge to maximum margin solutions in the feature space, leads the models to rely mostly on the simple features for classification. To mitigate this bias, we propose \emph{Feature Reconstruction Regularizer (FRR)} to ensure that the learned features can be reconstructed back from the logits. The use of \emph{FRR} in linear layer training (\emph{FRR-L}) encourages the use of more diverse features for classification. We further propose to finetune the full network by freezing the weights of the linear layer trained using \emph{FRR-L}, to refine the learned features, making them more suitable for classification. Using this simple solution, we demonstrate up to 15\% gains in OOD accuracy on the recently introduced semi-synthetic datasets with extreme distribution shifts. Moreover, we demonstrate noteworthy gains over existing SOTA methods on the standard OOD benchmark DomainBed as well. | We propose a regularizer that enforces the reconstruction of features from the output logits of neural networks, in order to overcome Simplicity Bias and boost their OOD generalization. |
Fully-parametric language models generally require a huge number of model parameters to store the necessary knowledge for solving multiple natural language tasks in zero/few-shot settings. In addition, it is hard to adapt to the evolving world knowledge without the costly model re-training. In this paper, we develop a novel semi-parametric language model architecture, Knowledge-in-Context (KiC), which empowers a parametric text-to-text language model with a knowledge-rich external memory. Specifically, the external memory contains six different types of knowledge: entity, dictionary, commonsense, event, script, and causality knowledge. For each input instance, the KiC model adaptively selects a knowledge type and retrieves the most helpful pieces of knowledge. The input instance along with its knowledge augmentation is fed into a text-to-text model (e.g., T5) to generate the output answer, where both the input and the output are in natural language forms after prompting. Interestingly, we find that KiC can be identified as a special mixture-of-experts (MoE) model, where the knowledge selector plays the role of a router that is used to determine the sequence-to-expert assignment in MoE. This key observation inspires us to develop a novel algorithm for training KiC with an instance-adaptive knowledge selector. As a knowledge-rich semi-parametric language model, KiC only needs a much smaller parametric part to achieve superior zero-shot performance on unseen tasks. By evaluating on 40+ different tasks, we show that KiC-Large with 770M parameters easily outperforms large language models that are 4-39x larger. In addition, KiC also exhibits emergent abilities at a much smaller model scale compared to the fully-parametric models. | We develop a novel knowledge-rich semi-parametric model, KiC, that is able to achieve superior zeros-hot performance on unseen task with a much smaller model size. |
State space models have shown to be effective at modeling long range dependencies, specially on sequence classification tasks. In this work we focus on autoregressive sequence modeling over English books, Github source code and ArXiv mathematics articles. Based on recent developments around the effectiveness of gated activation functions, we propose a new layer named \textit{Gated State Space} (GSS) and show that it trains significantly faster than the diagonal version of S4 (i.e. DSS) on TPUs, is fairly competitive with several well-tuned Transformer-based baselines and exhibits zero-shot generalization to longer inputs while being straightforward to implement. Finally, we show that leveraging self-attention to model local dependencies improves the performance of GSS even further. | Explore and improve state space model family on long range language modeling tasks |
Recent years have witnessed the prosperity of pre-training graph neural networks (GNNs) for molecules. Typically, atom types as node attributes are randomly masked, and GNNs are then trained to predict masked types as in AttrMask \citep{hu2020strategies}, following the Masked Language Modeling (MLM) task of BERT~\citep{devlin2019bert}. However, unlike MLM with a large vocabulary, the AttrMask pre-training does not learn informative molecular representations due to small and unbalanced atom `vocabulary'. To amend this problem, we propose a variant of VQ-VAE~\citep{van2017neural} as a context-aware tokenizer to encode atom attributes into chemically meaningful discrete codes. This can enlarge the atom vocabulary size and mitigate the quantitative divergence between dominant (e.g., carbons) and rare atoms (e.g., phosphorus). With the enlarged atom `vocabulary', we propose a novel node-level pre-training task, dubbed Masked Atoms Modeling (\textbf{MAM}), to mask some discrete codes randomly and then pre-train GNNs to predict them. MAM also mitigates another issue of AttrMask, namely the negative transfer. It can be easily combined with various pre-training tasks to improve their performance. Furthermore, we propose triplet masked contrastive learning (\textbf{TMCL}) for graph-level pre-training to model the heterogeneous semantic similarity between molecules for effective molecule retrieval. MAM and TMCL constitute a novel pre-training framework, \textbf{Mole-BERT}, which can match or outperform state-of-the-art methods in a fully data-driven manner. We release the code at \textcolor{magenta}{\url{https://github.com/junxia97/Mole-BERT}}. | We explain the negative transfer in molecular graph pre-training and develop two novel pre-training strategies to alleviate this issue. |
Test-time model adaptation has shown great effectiveness in generalizing over domain shifts. A most successful tactic for test-time adaptation conducts further optimization on the target data using the predictions by the source-trained model. However, due to domain shifts, the source-trained model predictions themselves can be largely inaccurate, which results in a model misspecified to the target data and therefore damages their adaptation ability. In this paper, we address test-time adaptation from a probabilistic perspective. We formulate model adaption as a probabilistic inference problem, which incorporates the uncertainty into source model predictions by modeling pseudo labels as distributions. Based on the probabilistic formalism, we propose variational pseudo labels that explore the information of neighboring target samples to improve pseudo labels and achieve a model better specified to target data. By a meta-learning paradigm, we train our model by simulating domain shifts and the test-time adaptation procedure. In doing so, our model learns the ability to generate more accurate pseudo-label distributions and to adapt to new domains. Experiments on three widely used datasets demonstrate the effectiveness of our proposal. | We address test-time adaptation in a probabilistic formulation by introducing variational pseudo labels with meta adaptation. |
Large-scale Vehicle Routing Problems (VRPs) are widely used in logistics, transportation, supply chain, and robotic systems. Recently, data-driven VRP heuristics are proposed to generate real-time VRP solutions with up to 100 nodes. Despite this progress, current heuristics for large-scale VRPs still face three major challenges: 1) Difficulty in generalizing the heuristics learned on small-scale VRPs to large-scale VRPs without retraining; 2) Challenge in generating real-time solutions for large-scale VRPs; 3) Difficulty in embedding global constraints into learned heuristics. We contribute in the three directions: We propose a Two-stage Divide Method (TAM) to generate sub-route sequence rather than node sequence for generalizing the heuristics learned on small-scale VRPs to solve large-scale VRPs in real-time. A two-step reinforcement learning method with new reward and padding techniques is proposed to train our TAM. A global mask function is proposed to keep the global constraints satisfied when dividing a large-scale VRP into several small-scale Traveling Salesman Problems (TSPs). As result, we can solve the small-scale TSPs in parallel quickly. The experiments on synthetic and real-world large-scale VRPs show our method could generalize the learned heuristics trained on datasets of VRP 100 to solve VRPs with over 5000 nodes in real-time while keeping the solution quality better than data-driven heuristics and competitive with traditional heuristics. | Propose a zero-shot method to generalize the data-driven heuristics trained on small-scale VRPs to solve large-scale VRPs in real-time |
Open-World Instance Segmentation (OWIS) is an emerging research topic that aims to segment class-agnostic object instances from images. The mainstream approaches use a two-stage segmentation framework, which first locates the candidate object bounding boxes and then performs instance segmentation. In this work, we instead promote a single-stage framework for OWIS. We argue that the end-to-end training process in the single-stage framework can be more convenient for directly regularizing the localization of class-agnostic object pixels. Based on the single-stage instance segmentation framework, we propose a regularization model to predict foreground pixels and use its relation to instance segmentation to construct a cross-task consistency loss. We show that such a consistency loss could alleviate the problem of incomplete instance annotation -- a common problem in the existing OWIS datasets. We also show that the proposed loss lends itself to an effective solution to semi-supervised OWIS that could be considered an extreme case that all object annotations are absent for some images. Our extensive experiments demonstrate that the proposed method achieves impressive results in both fully-supervised and semi-supervised settings. Compared to SOTA methods, the proposed method significantly improves the $AP_{100}$ score by 4.75\% in UVO$\rightarrow$UVO setting and 4.05\% in COCO$\rightarrow$UVO setting. In the case of semi-supervised learning, our model learned with only 30\% labeled data, even outperforms its fully-supervised counterpart with 50\% labeled data. The code will be released soon. | This paper proposed a single-stage open-world instance segmentation framework with a cross-task consistency loss, achieving superior performance. |
Nonlinear dynamics is ubiquitous in nature and commonly seen in various science and engineering disciplines. Distilling analytical expressions that govern nonlinear dynamics from limited data remains vital but challenging. To tackle this fundamental issue, we propose a novel Symbolic Physics Learner (SPL) machine to discover the mathematical structure of nonlinear dynamics. The key concept is to interpret mathematical operations and system state variables by computational rules and symbols, establish symbolic reasoning of mathematical formulas via expression trees, and employ a Monte Carlo tree search (MCTS) agent to explore optimal expression trees based on measurement data. The MCTS agent obtains an optimistic selection policy through the traversal of expression trees, featuring the one that maps to the arithmetic expression of underlying physics. Salient features of the proposed framework include search flexibility and enforcement of parsimony for discovered equations. The efficacy and superiority of the SPL machine are demonstrated by numerical examples, compared with state-of-the-art baselines. | Proposed a novel Symbolic Physics Learner (SPL) machine to discover the mathematical structure of nonlinear dynamics based on limited measurement data. |
Numerical reasoning over natural language has been a long-standing goal for the research community. However, recent language models have proven difficult to reliably generalize to a broad range of numbers, although they have shown proficiency in reasoning over common and simple numbers. In this paper, we propose a novel method to elicit and exploit the numerical reasoning knowledge hidden in pre-trained language models using simple anchor numbers. Concretely, we first leverage simple numbers as anchors to probe the implicitly inferred arithmetic expressions from language models, and then explicitly apply the expressions on complex numbers to get corresponding answers. To inversely elicit arithmetic expressions, we transform and formulate the task as an analytically solvable linear system. Experimental results on several numerical reasoning benchmarks demonstrate that our approach is highly effective. More importantly, our approach works in the inference phase without extra model training, making it highly portable and achieving significant and consistent performance benefits across a variety of language models in zero-shot, few-shot, and fine-tuning scenarios. | leverage simple numbers as "anchors" to probe the implicitly inferred arithmetic relationship from language models and then explicitly apply the relationship on complex numbers |
Deep learning has led to remarkable advances in computer vision. Even so, today’s best models are brittle when presented with variations that differ even slightly from those seen during training. Minor shifts in the pose, color, or illumination of an object can lead to catastrophic misclassifications. State-of-the art models struggle to understand how a set of variations can affect different objects. We propose a framework for instilling a notion of how objects vary in more realistic settings. Our approach applies the formalism of Lie groups to capture continuous transformations to improve models’ robustness to distributional shifts. We apply our framework on top of state-of-the-art self-supervised learning (SSL) models, finding that explicitly modeling transformations with Lie groups leads to substantial performance gains of greater than 10% for MAE on both known instances seen in typical poses now presented in new poses, and on unknown instances in any pose. We also apply our approach to ImageNet, finding that the Lie operator improves performance by almost 4%. These results demonstrate the promise of learning transformations to improve model robustness. | We explicitly model variation in data using Lie groups to improve self-supervised vision models' robustness to pose changes |
(Stochastic) bilevel optimization is a frequently encountered problem in machine learning with a wide range of applications such as meta-learning, hyper-parameter optimization, and reinforcement learning. Most of the existing studies on this problem only focused on analyzing the convergence or improving the convergence rate, while little effort has been devoted to understanding its generalization behaviors. In this paper, we conduct a thorough analysis on the generalization of first-order (gradient-based) methods for the bilevel optimization problem. We first establish a fundamental connection between algorithmic stability and generalization error in different forms and give a high probability generalization bound which improves the previous best one from $O(\sqrt{n})$ to $O(\log n)$, where $n$ is the sample size. We then provide the first stability bounds for the general case where both inner and outer level parameters are subject to continuous update, while existing work allows only the outer level parameter to be updated. Our analysis can be applied in various standard settings such as strongly-convex-strongly-convex (SC-SC), convex-convex (C-C), and nonconvex-nonconvex (NC-NC). Our analysis for the NC-NC setting can also be extended to a particular nonconvex-strongly-convex (NC-SC) setting that is commonly encountered in practice. Finally, we corroborate our theoretical analysis and demonstrate how iterations can affect the generalization error by experiments on meta-learning and hyper-parameter optimization. | Generalization bounds in different settings for single-timescale gradient-based method |
Recently, Rissanen et al., (2022) have presented a new type of diffusion process for generative modeling based on heat dissipation, or blurring, as an alternative to isotropic Gaussian diffusion. Here, we show that blurring can equivalently be defined through a Gaussian diffusion process with non-isotropic noise. In making this connection, we bridge the gap between inverse heat dissipation and denoising diffusion, and we shed light on the inductive bias that results from this modeling choice. Finally, we propose a generalized class of diffusion models that offers the best of both standard Gaussian denoising diffusion and inverse heat dissipation, which we call Blurring Diffusion Models. | We show that blurring can equivalently be defined through a Gaussian diffusion process with non-isotropic noise, bridging the gap between inverse heat dissipation and denoising diffusion |
Lipschitz regularized $f$-divergences are constructed by imposing a bound on the Lipschitz constant of the discriminator in the variational representation. These divergences interpolate between the Wasserstein metric and $f$-divergences and provide a flexible family of loss functions for non-absolutely continuous (e.g. empirical) distributions, possibly with heavy tails. We first construct Lipschitz regularized gradient flows on the space of probability measures based on these divergences. Examples of such gradient flows are Lipschitz regularized Fokker-Planck and porous medium partial differential equations (PDEs) for the Kullback-Leibler and $\alpha$-divergences, respectively. The regularization corresponds to imposing a Courant–Friedrichs–Lewy numerical stability condition on the PDEs. For empirical measures, the Lipschitz regularization on gradient flows induces a numerically stable transporter/discriminator particle algorithm, where the generative particles are transported along the gradient of the discriminator. The gradient structure leads to a regularized Fisher information which is the total kinetic energy of the particles and can be used to track the convergence of the algorithm. The Lipschitz regularized discriminator can be implemented via neural network spectral normalization and the particle algorithm generates approximate samples from possibly high-dimensional distributions known only from data. Notably, our particle algorithm can generate synthetic data even in small sample size regimes. A new data processing inequality for the regularized divergence allows us to combine our particle algorithm with representation learning, e.g. autoencoder architectures. The resulting particle algorithm in latent space yields markedly improved generative properties in terms of efficiency and quality of the synthetic samples. From a statistical mechanics perspective the encoding can be interpreted dynamically as learning a better mobility for the generative particles. | We construct gradient flows, in real and latent spaces, as a generative tool to evolve empirical distributions in terms of a particle algorithm. |
Ensuring fairness in Federated Learning (FL) systems, i.e. ensuring a satisfactory performance for all of the diverse clients in the systems, is an important and challenging problem. There are multiple fair FL algorithms in the literature, which have been relatively successful in providing fairness. However, these algorithms mostly emphasize on the loss functions of worst-off clients to improve their performance, which often results in the suppression of well-performing ones. As a consequence, the system's overall average performance is usually sacrificed for achieving fairness. Motivated by this and inspired by two well-known risk modeling methods in Finance, Mean-Variance and Mean-Semi-Variance, we propose and study two new fair FL algorithms, Variance Reduction (VRed) and Semi-Variance Reduction (Semi-VRed). VRed encourages equality between clients loss functions by penalizing their variance. In contrast, Semi-VRed penalizes the discrepancy of only the worst-off clients loss functions from the average loss. Through extensive experiments on multiple vision and language datasets, we show that, Semi-VRed achieves SoTA performance in scenarios with highly heterogeneous data distributions by improving both fairness and the system overall average performance at the same time. | We propose two new algorithms for fair federated learning based on variance and semi-variance regularization |
Regularization is almost {\it de rigueur} when designing autoencoders that are sparse and robust to noise. Given the recent surge of interest in machine learning problems involving non-Euclidean data, in this paper we address the regularization of autoencoders on curved spaces. We show that by ignoring the underlying geometry of the data and applying standard vector space regularization techniques, autoencoder performance can be severely degraded, or worse, training can fail to converge. Assuming that both the data space and latent space can be modeled as Riemannian manifolds, we show how to construct regularization terms in a coordinate-invariant way, and develop geometric generalizations of the denoising autoencoder and reconstruction contractive autoencoder such that the essential properties that enable the estimation of the derivative of the log-probability density are preserved. Drawing upon various non-Euclidean data sets, we show that our geometric autoencoder regularization techniques can have important performance advantages over vector-spaced methods while avoiding other breakdowns that can result from failing to account for the underlying geometry. | We propose geometrically regularized autoencoders for non-Euclidean data and discuss their various use cases. |
In this work, our objective is to adapt a Deep generative model trained on a large-scale source dataset to multiple target domains with scarce data. Specifically, we focus on adapting a pre-trained Generative Adversarial Network (GAN) to a target domain without re-training the generator. Our method draws the motivation from the fact that out-of-distribution samples can be `embedded' onto the latent space of a pre-trained source-GAN. We propose to train a small latent-generation network during the inference stage, each time a batch of target samples is to be generated. These target latent codes are fed to the source-generator to obtain novel target samples. Despite using the same small set of target samples and the source generator, multiple independent training episodes of the latent-generation network results in the diversity of the generated target samples. Our method, albeit simple, can be used to generate data from multiple target distributions using a generator trained on a single source distribution. We demonstrate the efficacy of our surprisingly simple method in generating multiple target datasets with only a single source generator and a few target samples. | Adapt a GAN trained on a single large-scale source dataset to multiple target domains containing very few examples without re-training the pretrained source generator. |
Continuous-time (CT) modeling has proven to provide improved sample efficiency and interpretability in learning the dynamical behavior of physical systems compared to discrete-time (DT) models. However, even with numerous recent developments, the CT nonlinear state-space (NL-SS) model identification problem remains to be solved in full, considering common experimental aspects such as the presence of external inputs, measurement noise, latent states, and general robustness. This paper presents a novel estimation method that addresses all these aspects and that can obtain state-of-the-art results on multiple benchmarks with compact fully connected neural networks capturing the CT dynamics. The proposed estimation method called the subspace encoder approach (SUBNET) ascertains these results by efficiently approximating the complete simulation loss by evaluating short simulations on subsections of the data, by using an encoder function to estimate the initial state for each subsection and a novel state-derivative normalization to ensure stability and good numerical conditioning of the training process. We prove that the use of subsections increases cost function smoothness together with the necessary requirements for the existence of the encoder function and we show that the proposed state-derivative normalization is essential for reliable estimation of CT NL-SS models. | This work proposes a method for the estimation of continuous-time nonlinear state-pace models parameterized by ANN's which is robust and well theoretically motivated. |
We consider a setting that a model needs to adapt to a new domain under distribution shifts, given that only unlabeled test samples from the new domain are accessible at test time. A common idea in most of the related works is constructing pseudo-labels for the unlabeled test samples and applying gradient descent (GD) to a loss function with the pseudo-labels. Recently, Goyal et al. (2022) propose conjugate labels, which is a new kind of pseudo-labels for self-training at test time. They empirically show that the conjugate label outperforms other ways of pseudo-labeling on many domain adaptation benchmarks. However, provably showing that GD with conjugate labels learns a good classifier for test-time adaptation remains open. In this work, we aim at theoretically understanding GD with hard and conjugate labels for a binary classification problem. We show that for square loss, GD with conjugate labels converges to an $\epsilon$-optimal predictor under a Gaussian model for any arbitrarily small $\epsilon$, while GD with hard pseudo-labels fails in this task. We also analyze them under different loss functions for the update. Our results shed lights on understanding when and why GD with hard labels or conjugate labels works in test-time adaptation. | Towards Understanding GD with Hard and Conjugate Pseudo-labels for Test-Time Adaptation |
In many real-world settings, image observations of freely rotating 3D rigid bodies, such as satellites, may be available when low-dimensional measurements are not. However, the high-dimensionality of image data precludes the use of classical estimation techniques to learn the dynamics and a lack of interpretability reduces the usefulness of standard deep learning methods. In this work, we present a physics-informed neural network model to estimate and predict 3D rotational dynamics from image sequences. We achieve this using a multi-stage prediction pipeline that maps individual images to a latent representation homeomorphic to $\mathbf{SO}(3)$, computes angular velocities from latent pairs, and predicts future latent states using the Hamiltonian equations of motion with a learned representation of the Hamiltonian. We demonstrate the efficacy of our approach on a new rotating rigid-body dataset with sequences of rotating cubes and rectangular prisms with uniform and non-uniform density. | Using Hamiltonian structure to learn interpretable dynamics from images of rotating 3D objects |
Existing pre-trained models are generally geared towards a particular class of problems. To date, there seems to be still no consensus on what the right architecture and pre-training setup should be. This paper presents a unified framework for pre-training models that are universally effective across datasets and setups. We begin by disentangling architectural archetypes with pre-training objectives -- two concepts that are commonly conflated. Next, we present a generalized and unified perspective for self-supervision in NLP and show how different pre-training objectives can be cast as one another and how interpolating between different objectives can be effective. We then propose Mixture-of-Denoisers (MoD), a pre-training objective that combines diverse pre-training paradigms together. We furthermore introduce a notion of mode switching, wherein downstream fine-tuning is associated with specific pre-training schemes. We conduct extensive ablative experiments to compare multiple pre-training objectives and find that our method pushes the Pareto-frontier by outperforming T5 and/or GPT-like models across multiple diverse setups. Finally, by scaling our model up to 20B parameters, we achieve SOTA performance on 50 well-established supervised NLP tasks ranging from language generation (with automated and human evaluation), language understanding, text classification, question answering, commonsense reasoning, long text reasoning, structured knowledge grounding and information retrieval. Our model also achieve strong results at in-context learning, outperforming 175B GPT-3 on zero-shot SuperGLUE and tripling the performance of T5-XXL on one-shot summarization. Finally, we show that UL2 20B works well with chain-of-thought prompting and reasoning, making it an appealing choice for research into reasoning at a small to medium scale of 20B parameters. We release Flax-based T5X model checkpoints for the 20B model publicly.
| How to train a language model properly |
Multivariate Time Series forecasting has been an increasingly popular topic in various applications and scenarios. Recently, contrastive learning and Transformer-based models have achieved good performance in many long-term series forecasting tasks. However, there are still several issues in existing methods. First, the training paradigm of contrastive learning and downstream prediction tasks are inconsistent, leading the accuracy of prediction not good enough. Second, existing Transformer-based models which learn similar patterns in historical time series data to predict future values always induces severe distribution shift problems, and does not fully leverage the sequence information compared to self-supervised methods. To address these issues, we propose a novel framework named Ti-MAE, in which the input time series are assumed to follow an integrate distribution. In detail, Ti-MAE randomly masks out embedded time series data and learns an autoencoder to reconstruct them at the point-level. Ti-MAE adopts mask modeling as the auxiliary task rather than contrastive learning and bridges the connection between existing representation learning and generative Transformer-based methods, reducing the difference between upstream and downstream forecasting tasks while maintaining the utilization of original time series data. Experiments on several public real-world datasets demonstrate that our framework of masked autoencoding could learn strong representations directly from the raw data, yielding better performance in time series forecasting and classification tasks. The code will be made public after this paper is accepted. | Self-Supervised Masked Time Series Autoencoders for time series forecasting and classification |
Deep Reinforcement Learning (Deep RL) and Evolutionary Algorithm (EA) are two major paradigms of policy optimization with distinct learning principles, i.e., gradient-based v.s. gradient-free. An appealing research direction is integrating Deep RL and EA to devise new methods by fusing their complementary advantages. However, existing works on combining Deep RL and EA have two common drawbacks:1) the RL agent and EA agents learn their policies individually, neglecting efficient sharing of useful common knowledge; 2) parameter-level policy optimization guarantees no semantic level of behavior evolution for the EA side. In this paper, we propose Evolutionary Reinforcement Learning with Two-scale State Representation and Policy Representation (ERL-Re$^2$), a novel solution to the aforementioned two drawbacks. The key idea of ERL-Re$^2$ is two-scale representation: all EA and RL policies share the same nonlinear state representation while maintaining individual linear policy representations. The state representation conveys expressive common features of the environment learned by all the agents collectively; the linear policy representation provides a favorable space for efficient policy optimization, where novel behavior-level crossover and mutation operations can be performed. Moreover, the linear policy representation allows convenient generalization of policy fitness with the help of Policy-extended Value Function Approximator (PeVFA), further improving the sample efficiency of fitness estimation. The experiments on a range of continuous control tasks show that ERL-Re$^2$ consistently outperforms advanced baselines and achieves the State Of The Art (SOTA). Our code is available on https://github.com/yeshenpy/ERL-Re2. | A novel and effective framework to fuse Reinforcement Learning and Evolutionary Algorithm for policy optimization. |
Adversarial training (AT) with samples generated by Fast Gradient Sign Method (FGSM), also known as FGSM-AT, is a computationally simple method to train robust networks. However, during its training procedure, an unstable mode of ``catastrophic overfitting'' has been identified in [Wong2020FastIB], where the robust accuracy abruptly drops to zero within a single training step. Existing methods use gradient regularizers or random initialization tricks to attenuate this issue, whereas they either take high computational cost or lead to lower robust accuracy. In this work, we provide the first study, which thoroughly examines a collection of tricks from three perspectives: Data Initialization, Network Structure, and Optimization, to overcome the catastrophic overfitting in FGSM-AT.
Surprisingly, we find that simple tricks, i.e., a) masking partial pixels (even without randomness), b) setting a large convolution stride and smooth activation functions, or c) regularizing the weights of the first convolutional layer, can effectively tackle the overfitting issue.
Extensive results on a range of network architectures validate the effectiveness of each proposed trick, and the combinations of tricks are also investigated. For example, trained with PreActResNet-18 on CIFAR-10, our method attains 49.8% accuracy against PGD-50 attacker and 46.4% accuracy against AutoAttack, demonstrating that pure FGSM-AT is capable of enabling robust learners. | We provide the first study, which thoroughly examines a collection of tricks, to overcome the catastrophic overfitting in FGSM-AT. |
Open set detection (OSD) aims at identifying data samples of an unknown class ($i.e.$, open set) from those of known classes ($i.e.$, closed set) based on a model trained from closed set samples. However, a closed set may involve a highly imbalanced class distribution. Accurately differentiating open set samples and those from a minority class in the closed set poses a fundamental challenge as the model may be equally uncertain when recognizing samples from the minority class. In this paper, we propose Adaptive Robust Evidential Optimization (AREO) that offers a principled way to quantify sample uncertainty through evidential learning while optimally balancing the model training over all classes in the closed set through adaptive distributively robust optimization (DRO). To avoid the model to primarily focus on the most difficult samples by following the standard DRO, adaptive DRO training is performed, which is governed by a novel multi-scheduler learning mechanism to ensure an optimal model training behavior that gives sufficient attention to the difficult samples and the minority class while capable of learning common patterns from the majority classes. Our experimental results on multiple real-world datasets demonstrate that the proposed model outputs uncertainty scores that can clearly separate samples from closed and open sets, respectively, and the detection results outperform the competitive baselines. | We propose adaptive robust uncertainty mass quantification for effective open set detection from imbalanced data. |
Reward and representation learning are two long-standing challenges for learning an expanding set of robot manipulation skills from sensory observations. Given the inherent cost and scarcity of in-domain, task-specific robot data, learning from large, diverse, offline human videos has emerged as a promising path towards acquiring a generally useful visual representation for control; however, how these human videos can be used for general-purpose reward learning remains an open question. We introduce $\textbf{V}$alue-$\textbf{I}$mplicit $\textbf{P}$re-training (VIP), a self-supervised pre-trained visual representation capable of generating dense and smooth reward functions for unseen robotic tasks. VIP casts representation learning from human videos as an offline goal-conditioned reinforcement learning problem and derives a self-supervised dual goal-conditioned value-function objective that does not depend on actions, enabling pre-training on unlabeled human videos. Theoretically, VIP can be understood as a novel implicit time contrastive objective that generates a temporally smooth embedding, enabling the value function to be implicitly defined via the embedding distance, which can then be used to construct the reward for any goal-image specified downstream task. Trained on large-scale Ego4D human videos and without any fine-tuning on in-domain, task-specific data, VIP can provide dense visual reward for an extensive set of simulated and $\textbf{real-robot}$ tasks, enabling diverse reward-based visual control methods and significantly outperforming all prior pre-trained representations. Notably, VIP can enable simple, few-shot offline RL on a suite of real-world robot tasks with as few as 20 trajectories. | A method for pre-training a goal-conditioned value function on human videos that can be effectively used as zero-shot visual reward and representation for unseen robotics tasks in simulation and real-world. |
Recently, contrastive learning approaches (e.g., CLIP (Radford et al., 2021)) have received huge success in multimodal learning, where the model tries to minimize the distance between the representations of different views (e.g., image and its caption) of the same data point, while keep the representations of different data points away from each other. However, from a theoretical perspective, it is unclear how contrastive learning can learn to align the representations from different views efficiently, especially in cases where the data is not isotropic. In this work, we analyze the training dynamics of a simple multimodal contrastive learning model, and show that contrastive pairs are important for the model to efficiently balance the learned representations. In particular, we reveal a stage-wise behavior of the learning process: In the first stage, the model aligns the feature representations using positive pairs and the condition number grows in this stage. Then, in the second stage, the model reduces the condition number of the learned representations using negative pairs. | We show that contrastive pairs are important for models to learn aligned and balanced representations in multimodal learning. |
Applying reinforcement learning to combinatorial optimization problems is attractive as it obviates the need for expert knowledge or pre-solved instances. However, it is unrealistic to expect an agent to solve these (often NP-)hard problems in a single shot at inference due to their inherent complexity, thus leading approaches are often augmented with additional search strategies, from stochastic sampling and beam-search to explicit fine-tuning.
In this paper, we argue for the benefits of learning a population of complementary agents, which can be simultaneously rolled out at inference. To this end, we introduce Poppy, a simple theoretically grounded training procedure for populations. Instead of relying on a predefined or hand-crafted notion of diversity, Poppy induces an unsupervised specialization targeted solely at maximizing the performance of the whole population. We show that Poppy leads to a set of complementary heuristics, and obtain state-of-the-art results on three popular NP-hard problems: the traveling salesman (TSP), the capacitated vehicle routing (CVRP), and 0-1 knapsack (KP). On TSP specifically, Poppy divides by 5 the optimality gap while reducing the inference time by more than 10 compared to previous state-of-the-art reinforcement learning approaches. | We present a population-based RL method for CO problems: the training procedure makes the agents complementary to maximize the population's performance. |
Evaluating clustering quality with reliable evaluation metrics like normalized mutual information (NMI) requires labeled data that can be expensive to annotate. We focus on the underexplored problem of estimating clustering quality with limited labels. We adapt existing approaches from the few-sample model evaluation literature to actively sub-sample, with a learned surrogate model, the most informative data points for annotation to estimate the evaluation metric. However, we find that their estimation can be biased and only relies on the labeled data. To that end, we introduce CEREAL, a comprehensive framework for few-sample clustering evaluation that extends active sampling approaches in three key ways. First, we propose novel NMI-based acquisition functions that account for the distinctive properties of clustering and uncertainties from a learned surrogate model. Next, we use ideas from semi-supervised learning and train the surrogate model with both the labeled and unlabeled data. Finally, we pseudo-label the unlabeled data with the surrogate model. We run experiments to estimate NMI in an active sampling pipeline on three datasets across vision and language. Our results show that CEREAL reduces the area under the absolute error curve by up to 57% compared to the best sampling baseline. We perform an extensive ablation study to show that our framework is agnostic to the choice of clustering algorithm and evaluation metric. We also extend CEREAL from clusterwise annotations to pairwise annotations. Overall, CEREAL can efficiently evaluate clustering with limited human annotations. | We introduce CEREAL, a comprehensive framework for few-sample clustering evaluation based on active pseudo-labeling. |
Beyond the success story of adversarial training (AT) in the recent text domain on top of pre-trained language models (PLMs), our empirical study showcases the inconsistent gains from AT on some tasks, e.g. commonsense reasoning, named entity recognition. This paper investigates AT from the perspective of the contextualized language representation outputted by PLM encoders. We find the current AT attacks lean to generate sub-optimal adversarial examples that can fool the decoder part but have a minor effect on the encoder. However, we find it necessary to effectively deviate the latter one to allow AT to gain. Based on the observation, we propose simple yet effective \textit{Contextualized representation-Adversarial Training} (CreAT), in which the attack is explicitly optimized to deviate the contextualized representation of the encoder. It allows a global optimization of adversarial examples that can fool the entire model. We also find CreAT gives rise to a better direction to optimize the adversarial examples, to let them less sensitive to hyperparameters. Compared to AT, CreAT produces consistent performance gains on a wider range of tasks and is proven to be more effective for language pre-training where only the encoder part is kept for downstream tasks. We achieve the new state-of-the-art performances on a series of challenging benchmarks, e.g. AdvGLUE (59.1 $ \rightarrow $ 61.1), HellaSWAG (93.0 $ \rightarrow $ 94.9), ANLI (68.1 $ \rightarrow $ 69.3). | Adversarial training optimized to deviate contextualized language representation with all-powerful performance gain |
The training process of generative adversarial networks (GANs) is unstable and does not converge globally. In this paper, we propose a universal higher order noise based control called Brownian Motion Control (BMC) that is invariant to GANs frameworks so that the training process of GANs is exponential stable almost surely. Specifically, starting with the prototypical case of Dirac-GANs, we design a BMC and propose Dirac-BrGANs that retrieve exactly the same but reachable optimal equilibrium regardless of GANs' framework. The optimal equilibrium of our Dirac-BrGANs' training system is globally unique and always exists. Furthermore, the training process of Dirac-BrGANs achieve exponentially stability almost surely for any arbitrary initial value. Then we extend our BMC to normal GANs' settings and propose BrGANs. We provide numerical experiments showing that our BrGANs effectively stabilizes GANs's training process and obtains state-of-the art performance compared to other stabilizing methods. | We propose a higher order Brownian Motion Controller (BMC) for BrGANs to stablize GANs' training process |
Recent efforts to learn reward functions from human feedback have tended to use deep neural networks, whose lack of transparency hampers our ability to explain agent behaviour or verify alignment. We explore the merits of learning intrinsically interpretable tree models instead. We develop a recently proposed method for learning reward trees from preference labels, and show it to be broadly competitive with neural networks on challenging high-dimensional tasks, with good robustness to limited or corrupted data. Having found that reward tree learning can be done effectively in complex settings, we then consider why it should be used, demonstrating that the interpretable reward structure gives significant scope for traceability, verification and explanation. | We show that reward learning with tree models can be competitive with neural networks, and demonstrate some of its interpretability benefits. |
Detecting out-of-distribution (OOD) data is critical to building reliable machine learning systems in the open world. Previous works mainly perform OOD detection in feature space or output space. Recently, researchers have achieved promising results using gradient information, which combines the information in both feature and output space for OOD detection. However, existing works still suffer from the problem of overconfidence. To address this problem, we propose a novel method called ``Reactivate Gradnorm (RG)'', which exploits the norm of the clipped feature vector and the energy in the output space for OOD detection. To verify the effectiveness of our method, we conduct experiments on four benchmark datasets. Experimental results demonstrate that our RG outperforms existing state-of-the-art approaches by 2.06\% in average AUROC. Meanwhile, RG is easy to implement and does not require additional OOD data or fine-tuning process. We can realize OOD detection in only one forward pass of any pretrained model. | The information of joint feature space and output space improves the performance of OOD detection. |
In this paper, we investigate the problem of certifying neural network robustness against universal perturbations (UPs), which have been widely used in universal adversarial attacks and backdoor attacks. Existing robustness certification methods aim to provide robustness guarantees for each sample with respect to the worst-case perturbations given a neural network. However, those sample-wise bounds will be loose when considering the UP threat model as they overlook the important constraint that the perturbation should be shared across all samples. We propose a method based on a combination of linear relaxation-based perturbation analysis and Mixed Integer Linear Programming to establish the first robust certification method for UP. In addition, we develop a theoretical framework for computing error bounds on the entire population using the certification results from a randomly sampled batch. Aside from an extensive evaluation of the proposed certification, we further show how the certification facilitates efficient comparison of robustness among different models or efficacy among different universal adversarial attack defenses and enables accurate detection of backdoor target classes. | A robustness certification framework against universal perturbations (including both universal adversarial noise and backdoor attacks). |