Skip to yearly menu bar Skip to main content


Oral Session 12

Moderators: Dustin Tran · Taylor Berg-Kirkpatrick · Prateek Jain


Chat is not available.

Thu 6 May 19:00 - 19:15 PDT

Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data

Colin Wei · Kendrick Shen · Yining Chen · Tengyu Ma

Self-training algorithms, which train a model to fit pseudolabels predicted by another previously-learned model, have been very successful for learning with unlabeled data using neural networks. However, the current theoretical understanding of self-training only applies to linear models. This work provides a unified theoretical analysis of self-training with deep networks for semi-supervised learning, unsupervised domain adaptation, and unsupervised learning. At the core of our analysis is a simple but realistic “expansion” assumption, which states that a low-probability subset of the data must expand to a neighborhood with large probability relative to the subset. We also assume that neighborhoods of examples in different classes have minimal overlap. We prove that under these assumptions, the minimizers of population objectives based on self-training and input-consistency regularization will achieve high accuracy with respect to ground-truth labels. By using off-the-shelf generalization bounds, we immediately convert this result to sample complexity guarantees for neural nets that are polynomial in the margin and Lipschitzness. Our results help explain the empirical successes of recently proposed self-training algorithms which use input consistency regularization.

Thu 6 May 19:15 - 19:25 PDT

Long-tailed Recognition by Routing Diverse Distribution-Aware Experts

Xudong Wang · Long Lian · Zhongqi Miao · Ziwei Liu · Stella Yu

Natural data are often long-tail distributed over semantic classes. Existing recognition methods tend to focus on gaining performance on tail classes, often at the expense of losing performance on head classes and with increased classifier variance. The low tail performance manifests itself in large inter-class confusion and high classifier variance. We aim to reduce both the bias and the variance of a long-tailed classifier by RoutIng Diverse Experts (RIDE), consisting of three components: 1) a shared architecture for multiple classifiers (experts); 2) a distribution-aware diversity loss that encourages more diverse decisions for classes with fewer training instances; and 3) an expert routing module that dynamically assigns more ambiguous instances to additional experts. With on-par computational complexity, RIDE significantly outperforms the state-of-the-art methods by 5% to 7% on all the benchmarks including CIFAR100-LT, ImageNet-LT, and iNaturalist 2018. RIDE is also a universal framework that can be applied to different backbone networks and integrated into various long-tailed algorithms and training mechanisms for consistent performance gains. Our code is publicly available at

Thu 6 May 19:25 - 19:35 PDT

Self-Supervised Policy Adaptation during Deployment

Nicklas Hansen · Rishabh Jangir · Yu Sun · Guillem Alenyà · Pieter Abbeel · Alexei Efros · Lerrel Pinto · Xiaolong Wang

In most real world scenarios, a policy trained by reinforcement learning in one environment needs to be deployed in another, potentially quite different environment. However, generalization across different environments is known to be hard. A natural solution would be to keep training after deployment in the new environment, but this cannot be done if the new environment offers no reward signal. Our work explores the use of self-supervision to allow the policy to continue training after deployment without using any rewards. While previous methods explicitly anticipate changes in the new environment, we assume no prior knowledge of those changes yet still obtain significant improvements. Empirical evaluations are performed on diverse simulation environments from DeepMind Control suite and ViZDoom, as well as real robotic manipulation tasks in continuously changing environments, taking observations from an uncalibrated camera. Our method improves generalization in 31 out of 36 environments across various tasks and outperforms domain randomization on a majority of environments. Webpage and implementation:

Thu 6 May 19:35 - 19:45 PDT

What are the Statistical Limits of Offline RL with Linear Function Approximation?

Ruosong Wang · Dean Foster · Sham M Kakade

Offline reinforcement learning seeks to utilize offline (observational) data to guide the learning of (causal) sequential decision making strategies. The hope is that offline reinforcement learning coupled with function approximation methods (to deal with the curse of dimensionality) can provide a means to help alleviate the excessive sample complexity burden in modern sequential decision making problems. However, the extent to which this broader approach can be effective is not well understood, where the literature largely consists of sufficient conditions.

This work focuses on the basic question of what are necessary representational and distributional conditions that permit provable sample-efficient offline reinforcement learning. Perhaps surprisingly, our main result shows that even if: i) we have realizability in that the true value function of \emph{every} policy is linear in a given set of features and 2) our off-policy data has good coverage over all features (under a strong spectral condition), any algorithm still (information-theoretically) requires a number of offline samples that is exponential in the problem horizon to non-trivially estimate the value of \emph{any} given policy. Our results highlight that sample-efficient offline policy evaluation is not possible unless significantly stronger conditions hold; such conditions include either having low distribution shift (where the offline data distribution is close to the distribution of the policy to be evaluated) or significantly stronger representational conditions (beyond realizability).

Thu 6 May 19:45 - 19:55 PDT


Thu 6 May 19:55 - 20:05 PDT

RMSprop converges with proper hyper-parameter

Naichen Shi · Dawei Li · Mingyi Hong · Ruoyu Sun

Despite the existence of divergence examples, RMSprop remains one of the most popular algorithms in machine learning. Towards closing the gap between theory and practice, we prove that RMSprop converges with proper choice of hyper-parameters under certain conditions. More specifically, we prove that when the hyper-parameter $\beta_2$ is close enough to $1$, RMSprop and its random shuffling version converge to a bounded region in general, and to critical points in the interpolation regime. It is worth mentioning that our results do not depend on ``bounded gradient" assumption, which is often the key assumption utilized by existing theoretical work for Adam-type adaptive gradient method. Removing this assumption allows us to establish a phase transition from divergence to non-divergence for RMSprop. Finally, based on our theory, we conjecture that in practice there is a critical threshold $\sf{\beta_2^*}$, such that RMSprop generates reasonably good results only if $1>\beta_2\ge \sf{\beta_2^*}$. We provide empirical evidence for such a phase transition in our numerical experiments.

Thu 6 May 20:05 - 20:15 PDT

A Good Image Generator Is What You Need for High-Resolution Video Synthesis

Yu Tian · Jian Ren · Menglei Chai · Kyle Olszewski · Xi Peng · Dimitris Metaxas · Sergey Tulyakov

Image and video synthesis are closely related areas aiming at generating content from noise. While rapid progress has been demonstrated in improving image-based models to handle large resolutions, high-quality renderings, and wide variations in image content, achieving comparable video generation results remains problematic. We present a framework that leverages contemporary image generators to render high-resolution videos. We frame the video synthesis problem as discovering a trajectory in the latent space of a pre-trained and fixed image generator. Not only does such a framework render high-resolution videos, but it also is an order of magnitude more computationally efficient. We introduce a motion generator that discovers the desired trajectory, in which content and motion are disentangled. With such a representation, our framework allows for a broad range of applications, including content and motion manipulation. Furthermore, we introduce a new task, which we call cross-domain video synthesis, in which the image and motion generators are trained on disjoint datasets belonging to different domains. This allows for generating moving objects for which the desired video data is not available. Extensive experiments on various datasets demonstrate the advantages of our methods over existing video generation techniques. Code will be released at

Thu 6 May 20:15 - 20:25 PDT

Random Feature Attention

Hao Peng · Nikolaos Pappas · Dani Yogatama · Roy Schwartz · Noah Smith · Lingpeng Kong

Transformers are state-of-the-art models for a variety of sequence modeling tasks. At their core is an attention function which models pairwise interactions between the inputs at every timestep. While attention is powerful, it does not scale efficiently to long sequences due to its quadratic time and space complexity in the sequence length. We propose RFA, a linear time and space attention that uses random feature methods to approximate the softmax function, and explore its application in transformers. RFA can be used as a drop-in replacement for conventional softmax attention and offers a straightforward way of learning with recency bias through an optional gating mechanism. Experiments on language modeling and machine translation demonstrate that RFA achieves similar or better performance compared to strong transformer baselines. In the machine translation experiment, RFA decodes twice as fast as a vanilla transformer. Compared to existing efficient transformer variants, RFA is competitive in terms of both accuracy and efficiency on three long text classification datasets. Our analysis shows that RFA’s efficiency gains are especially notable on long sequences, suggesting that RFA will be particularly useful in tasks that require working with large inputs, fast decoding speed, or low memory footprints.

Thu 6 May 20:25 - 20:35 PDT

Learning with Feature-Dependent Label Noise: A Progressive Approach

Yikai Zhang · Songzhu Zheng · Pengxiang Wu · Mayank Goswami · Chao Chen

Label noise is frequently observed in real-world large-scale datasets. The noise is introduced due to a variety of reasons; it is heterogeneous and feature-dependent. Most existing approaches to handling noisy labels fall into two categories: they either assume an ideal feature-independent noise, or remain heuristic without theoretical guarantees. In this paper, we propose to target a new family of feature-dependent label noise, which is much more general than commonly used i.i.d. label noise and encompasses a broad spectrum of noise patterns. Focusing on this general noise family, we propose a progressive label correction algorithm that iteratively corrects labels and refines the model. We provide theoretical guarantees showing that for a wide variety of (unknown) noise patterns, a classifier trained with this strategy converges to be consistent with the Bayes classifier. In experiments, our method outperforms SOTA baselines and is robust to various noise types and levels.

Thu 6 May 20:35 - 20:45 PDT

Sparse Quantized Spectral Clustering

Zhenyu Liao · Romain Couillet · Michael W Mahoney

Given a large data matrix, sparsifying, quantizing, and/or performing other entry-wise nonlinear operations can have numerous benefits, ranging from speeding up iterative algorithms for core numerical linear algebra problems to providing nonlinear filters to design state-of-the-art neural network models. Here, we exploit tools from random matrix theory to make precise statements about how the eigenspectrum of a matrix changes under such nonlinear transformations. In particular, we show that very little change occurs in the informative eigenstructure, even under drastic sparsification/quantization, and consequently that very little downstream performance loss occurs when working with very aggressively sparsified or quantized spectral clustering problems. We illustrate how these results depend on the nonlinearity, we characterize a phase transition beyond which spectral clustering becomes possible, and we show when such nonlinear transformations can introduce spurious non-informative eigenvectors.

Thu 6 May 20:45 - 20:58 PDT


Thu 6 May 20:58 - 21:08 PDT

Learning a Latent Simplex in Input Sparsity Time

Ainesh Bakshi · Chiranjib Bhattacharyya · Ravi Kannan · David Woodruff · Samson Zhou

We consider the problem of learning a latent $k$-vertex simplex $K\in\mathbb{R}^d$, given $\mathbf{A}\in\mathbb{R}^{d\times n}$, which can be viewed as $n$ data points that are formed by randomly perturbing some latent points in $K$, possibly beyond $K$. A large class of latent variable models, such as adversarial clustering, mixed membership stochastic block models, and topic models can be cast in this view of learning a latent simplex. Bhattacharyya and Kannan (SODA 2020) give an algorithm for learning such a $k$-vertex latent simplex in time roughly $O(k\cdot\text{nnz}(\mathbf{A}))$, where $\text{nnz}(\mathbf{A})$ is the number of non-zeros in $\mathbf{A}$. We show that the dependence on $k$ in the running time is unnecessary given a natural assumption about the mass of the top $k$ singular values of $\mathbf{A}$, which holds in many of these applications. Further, we show this assumption is necessary, as otherwise an algorithm for learning a latent simplex would imply a better low rank approximation algorithm than what is known. We obtain a spectral low-rank approximation to $\mathbf{A}$ in input-sparsity time and show that the column space thus obtained has small $\sin\Theta$ (angular) distance to the right top-$k$ singular space of $\mathbf{A}$. Our algorithm then selects $k$ points in the low-rank subspace with the largest inner product (in absolute value) with $k$ carefully chosen random vectors. By working in the low-rank subspace, we avoid reading the entire matrix in each iteration and thus circumvent the $\Theta(k\cdot\text{nnz}(\mathbf{A}))$ running time.

Thu 6 May 21:08 - 21:18 PDT

Topology-Aware Segmentation Using Discrete Morse Theory

Xiaoling Hu · Yusu Wang · Li Fuxin · Dimitris Samaras · Chao Chen

In the segmentation of fine-scale structures from natural and biomedical images, per-pixel accuracy is not the only metric of concern. Topological correctness, such as vessel connectivity and membrane closure, is crucial for downstream analysis tasks. In this paper, we propose a new approach to train deep image segmentation networks for better topological accuracy. In particular, leveraging the power of discrete Morse theory (DMT), we identify global structures, including 1D skeletons and 2D patches, which are important for topological accuracy. Trained with a novel loss based on these global structures, the network performance is significantly improved especially near topologically challenging locations (such as weak spots of connections and membranes). On diverse datasets, our method achieves superior performance on both the DICE score and topological metrics.

Thu 6 May 21:18 - 21:28 PDT

MARS: Markov Molecular Sampling for Multi-objective Drug Discovery

Yutong Xie · Chence Shi · Hao Zhou · Yuwei Yang · Weinan Zhang · Yong Yu · Lei Li

Searching for novel molecules with desired chemical properties is crucial in drug discovery. Existing work focuses on developing neural models to generate either molecular sequences or chemical graphs. However, it remains a big challenge to find novel and diverse compounds satisfying several properties. In this paper, we propose MARS, a method for multi-objective drug molecule discovery. MARS is based on the idea of generating the chemical candidates by iteratively editing fragments of molecular graphs. To search for high-quality candidates, it employs Markov chain Monte Carlo sampling (MCMC) on molecules with an annealing scheme and an adaptive proposal. To further improve sample efficiency, MARS uses a graph neural network (GNN) to represent and select candidate edits, where the GNN is trained on-the-fly with samples from MCMC. Experiments show that MARS achieves state-of-the-art performance in various multi-objective settings where molecular bio-activity, drug-likeness, and synthesizability are considered. Remarkably, in the most challenging setting where all four objectives are simultaneously optimized, our approach outperforms previous methods significantly in comprehensive evaluations. The code is available at

Thu 6 May 21:28 - 21:38 PDT

Distributional Sliced-Wasserstein and Applications to Generative Modeling

Khai Nguyen · Nhat Ho · Tung Pham · Hung Bui

Sliced-Wasserstein distance (SW) and its variant, Max Sliced-Wasserstein distance (Max-SW), have been used widely in the recent years due to their fast computation and scalability even when the probability measures lie in a very high dimensional space. However, SW requires many unnecessary projection samples to approximate its value while Max-SW only uses the most important projection, which ignores the information of other useful directions. In order to account for these weaknesses, we propose a novel distance, named Distributional Sliced-Wasserstein distance (DSW), that finds an optimal distribution over projections that can balance between exploring distinctive projecting directions and the informativeness of projections themselves. We show that the DSW is a generalization of Max-SW, and it can be computed efficiently by searching for the optimal push-forward measure over a set of probability measures over the unit sphere satisfying certain regularizing constraints that favor distinct directions. Finally, we conduct extensive experiments with large-scale datasets to demonstrate the favorable performances of the proposed distances over the previous sliced-based distances in generative modeling applications.

Thu 6 May 21:38 - 21:48 PDT