ICLR 2023 论文列表

Multi-view image compression plays a critical role in 3D-related applications. Existing methods adopt a predictive coding architecture, which requires joint encoding to compress the corresponding disparity as well as residual information. This demands collaboration among cameras and enforces the epipolar geometric constraint between different views, which makes it challenging to deploy these methods in distributed camera systems with randomly overlapping fields of view. Meanwhile, distributed source coding theory indicates that efficient data compression of correlated sources can be achieved by independent encoding and joint decoding, which motivates us to design a learning-based distributed multi-view image coding (LDMIC) framework. With independent encoders, LDMIC introduces a simple yet effective joint context transfer module based on the cross-attention mechanism at the decoder to effectively capture the global inter-view correlations, which is insensitive to the geometric relationships between images. Experimental results show that LDMIC significantly outperforms both traditional and learning-based MIC methods while enjoying fast encoding speed. Code is released at https://github.com/Xinjie-Q/LDMIC.

Random-feature-based attention (RFA) is an efficient approximation of softmax attention with linear runtime and space complexity. However, the approximation gap between RFA and conventional softmax attention is not well studied. Built upon previous progress of RFA, we characterize this gap through the lens of control variates and show that RFA can be decomposed into a sum of multiple control variate estimators for each element in the sequence. This new framework reveals that exact softmax attention can be recovered from RFA by manipulating each control variate. Besides, it allows us to develop a more flexible form of control variates, resulting in a novel attention mechanism that significantly reduces the approximation gap while maintaining linear complexity. Extensive experiments demonstrate that our model outperforms state-of-the-art efficient attention mechanisms on both vision and language tasks.

Discontinuous solutions arise in a large class of hyperbolic and advection-dominated PDEs. This paper investigates, both theoretically and empirically, the operator learning of PDEs with discontinuous solutions. We rigorously prove, in terms of lower approximation bounds, that methods which entail a linear reconstruction step (e.g. DeepONets or PCA-Nets) fail to efficiently approximate the solution operator of such PDEs. In contrast, we show that certain methods employing a non-linear reconstruction mechanism can overcome these fundamental lower bounds and approximate the underlying operator efficiently. The latter class includes Fourier Neural Operators and a novel extension of DeepONets termed shift-DeepONets. Our theoretical findings are confirmed by empirical results for advection equations, inviscid Burgers’ equation and the compressible Euler equations of gas dynamics.

The well-designed structures in neural networks reflect the prior knowledge incorporated into the models. However, though different models have various priors, we are used to training them with model-agnostic optimizers such as SGD. In this paper, we propose to incorporate model-specific prior knowledge into optimizers by modifying the gradients according to a set of model-specific hyper-parameters. Such a methodology is referred to as Gradient Re-parameterization, and the optimizers are named RepOptimizers. For the extreme simplicity of model structure, we focus on a VGG-style plain model and showcase that such a simple model trained with a RepOptimizer, which is referred to as RepOpt-VGG, performs on par with or better than the recent well-designed models. From a practical perspective, RepOpt-VGG is a favorable base model because of its simple structure, high inference speed and training efficiency. Compared to Structural Re-parameterization, which adds priors into models via constructing extra training-time structures, RepOptimizers require no extra forward/backward computations and solve the problem of quantization. We hope to spark further research beyond the realms of model structure design. Code and models https://github.com/DingXiaoH/RepOptimizers.

This paper novelly breaks down with ignorable loss an RNN layer into a sequence of simple RNNs, each of which can be further rewritten into a lightweight positional encoding matrix of a self-attention, named the Recurrence Encoding Matrix (REM). Thus, recurrent dynamics introduced by the RNN layer can be encapsulated into the positional encodings of a multihead self-attention, and this makes it possible to seamlessly incorporate these recurrent dynamics into a Transformer, leading to a new module, Self-Attention with Recurrence (RSA). The proposed module can leverage the recurrent inductive bias of REMs to achieve a better sample efficiency than its corresponding baseline Transformer, while the self-attention is used to model the remaining non-recurrent signals. The relative proportions of these two components are controlled by a data-driven gated mechanism, and the effectiveness of RSA modules are demonstrated by four sequential learning tasks.

Sharpness-Aware Minimization (SAM) is a highly effective regularization technique for improving the generalization of deep neural networks for various settings. However, the underlying working of SAM remains elusive because of various intriguing approximations in the theoretical characterizations. SAM intends to penalize a notion of sharpness of the model but implements a computationally efficient variant; moreover, a third notion of sharpness was used for proving generalization guarantees. The subtle differences in these notions of sharpness can indeed lead to significantly different empirical results. This paper rigorously nails down the exact sharpness notion that SAM regularizes and clarifies the underlying mechanism. We also show that the two steps of approximations in the original motivation of SAM individually lead to inaccurate local conclusions, but their combination accidentally reveals the correct effect, when full-batch gradients are applied. Furthermore, we also prove that the stochastic version of SAM in fact regularizes the third notion of sharpness mentioned above, which is most likely to be the preferred notion for practical performance. The key mechanism behind this intriguing phenomenon is the alignment between the gradient and the top eigenvector of Hessian when SAM is applied.

Recently the SP (Stochastic Polyak step size) method has emerged as a competitive adaptive method for setting the step sizes of SGD. SP can be interpreted as a method specialized to interpolated models, since it solves the interpolation equations. SP solves these equation by using local linearizations of the model. We take a step further and develop a method for solving the interpolation equations that uses the local second-order approximation of the model. Our resulting method SP2 uses Hessian-vector products to speed-up the convergence of SP. Furthermore, and rather uniquely among second-order methods, the design of SP2 in no way relies on positive definite Hessian matrices or convexity of the objective function. We show SP2 is competitive both in experiments and in theory. We show SP2 is very competitive on matrix completion, non-convex test problems and logistic regression. We also provide a convergence theory on sums-of-quadratics.

Shape-based virtual screening is widely used in ligand-based drug design to search chemical libraries for molecules with similar 3D shapes yet novel 2D graph structures compared to known ligands. 3D deep generative models can potentially automate this exploration of shape-conditioned 3D chemical space; however, no existing models can reliably generate geometrically realistic drug-like molecules in conformations with a specific shape. We introduce a new multimodal 3D generative model that enables shape-conditioned 3D molecular design by equivariantly encoding molecular shape and variationally encoding chemical identity. We ensure local geometric and chemical validity of generated molecules by using autoregressive fragment-based generation with heuristic bonding geometries, allowing the model to prioritize the scoring of rotatable bonds to best align the growing conformation to the target shape. We evaluate our 3D generative model in tasks relevant to drug design including shape-conditioned generation of chemically diverse molecular structures and shape-constrained molecular property optimization, demonstrating its utility over virtual screening of enumerated libraries.

Prevention of complete and dimensional collapse of representations has recently become a design principle for self-supervised learning (SSL). However, questions remain in our theoretical understanding: When do those collapses occur? What are the mechanisms and causes? We answer these questions by deriving and thoroughly analyzing an analytically tractable theory of SSL loss landscapes. In this theory, we identify the causes of the dimensional collapse and study the effect of normalization and bias. Finally, we leverage the interpretability afforded by the analytical theory to understand how dimensional collapse can be beneficial and what affects the robustness of SSL against data imbalance.

There has been a debate on the choice of plain vs. hierarchical vision transformers, where researchers often believe that the former (e.g., ViT) has a simpler design but the latter (e.g., Swin) enjoys higher recognition accuracy. Recently, the emerge of masked image modeling (MIM), a self-supervised visual pre-training method, raised a new challenge to vision transformers in terms of flexibility, i.e., part of image patches or tokens are to be discarded, which seems to claim the advantages of plain vision transformers. In this paper, we delve deep into the comparison between ViT and Swin, revealing that (i) the performance gain of Swin is mainly brought by a deepened backbone and relative positional encoding, (ii) the hierarchical design of Swin can be simplified into hierarchical patch embedding (proposed in this work), and (iii) other designs such as shifted-window attentions can be removed. By removing the unnecessary operations, we come up with a new architecture named HiViT (short for hierarchical ViT), which is simpler and more efficient than Swin yet further improves its performance on fully-supervised and self-supervised visual representation learning. In particular, after pre-trained using masked autoencoder (MAE) on ImageNet-1K, HiViT-B reports a 84.6% accuracy on ImageNet-1K classification, a 53.3% box AP on COCO detection, and a 52.8% mIoU on ADE20K segmentation, significantly surpassing the baseline. Code is available at https://github.com/zhangxiaosong18/hivit.

Neural networks that are equivariant to rotations, translations, reflections, and permutations on $n$-dimensional geometric space have shown promise in physical modeling for tasks such as accurately but inexpensively modeling complex potential energy surfaces to guiding the sampling of complex dynamical systems or forecasting their time evolution. Current state-of-the-art methods employ spherical harmonics to encode higher-order interactions among particles, which are computationally expensive. In this paper, we propose a simple alternative functional form that uses neurally parametrized linear combinations of edge vectors to achieve equivariance while still universally approximating node environments. Incorporating this insight, we design \emph{spatial attention kinetic networks} with E(n)-equivariance, or SAKE, which are competitive in many-body system modeling tasks while being significantly faster.

Out of distribution (OoD) generalization has received considerable interest in recent years. In this work, we identify a particular failure mode of OoD generalization for discriminative classifiers that is based on test data (from a new domain) lying in the nullspace of features learnt from source data. We demonstrate the existence of this failure mode across multiple networks trained across RotatedMNIST, PACS, TerraIncognita, DomainNet and ImageNet-R datasets. We then study different choices for characterizing the feature space and show that projecting intermediate representations onto the span of directions that obtain maximum training accuracy provides consistent improvements in OoD performance. Finally, we show that such nullspace behavior also provides an insight into neural networks trained on poisoned data. We hope our work galvanizes interest in the relationship between the nullspace occupancy failure mode and generalization.

Reward design in reinforcement learning (RL) is challenging since specifying human notions of desired behavior may be difficult via reward functions or require many expert demonstrations. Can we instead cheaply design rewards using a natural language interface? This paper explores how to simplify reward design by using a large language model (LLM) such as GPT-3 as a proxy reward function, where the user provides a textual prompt containing a few examples (few-shot) or a description (zero-shot) of desired behavior. Our approach leverages this proxy reward function in an RL framework. Specifically, users specify a prompt once at the beginning of training. During training, the LLM evaluates an RL agent's behavior against the desired behavior described by the prompt and outputs a corresponding reward signal. The RL agent then uses this reward to update its behavior. We evaluate whether our approach can train agents aligned with user objectives in the Ultimatum Game, matrix games, and the DealOrNoDeal negotiation task. In all three tasks, we show that RL agents trained with our framework are well-aligned with the user's objectives and outperforms RL agents trained with reward functions learned via supervised learning.