Search Results for author:
Found 102 papers, 13 papers with code
Lower Complexity Bounds for Finite-Sum Convex-Concave Minimax Optimization Problems
This paper studies the lower bound complexity for minimax optimization problem whose objective function is the average of $n$ individual smooth convex-concave functions.
MR-P: A Parallel Decoding Algorithm for Iterative Refinement Non-Autoregressive Translation
The Conditional Masked Language Model (CMLM) is a strong baseline of NAT.
Near Minimax-Optimal Distributional Temporal Difference Algorithms and The Freedman Inequality in Hilbert Spaces
In the tabular case, \citet{rowland2018analysis} and \citet{rowland2023analysis} proved the asymptotic convergence of two instances of distributional TD, namely categorical temporal difference algorithm (CTD) and quantile temporal difference algorithm (QTD), respectively.
Finite-Time Decoupled Convergence in Nonlinear Two-Time-Scale Stochastic Approximation
In two-time-scale stochastic approximation (SA), two iterates are updated at varying speeds using different step sizes, with each update influencing the other.
Purify++: Improving Diffusion-Purification with Advanced Diffusion Models and Control of Randomness
Based on our findings, we propose Purify++, a new diffusion purification algorithm that is now the state-of-the-art purification method against several adversarial attacks.
Estimation and Inference in Distributional Reinforcement Learning
This implies the distributional policy evaluation problem can be solved with sample efficiency.
Distributional Reinforcement Learning
reinforcement-learning
Constructing Synthetic Treatment Groups without the Mean Exchangeability Assumption
Motivated by the synthetic control method, we construct a synthetic treatment group for the target population by a weighted mixture of treatment groups of source populations.
Training Energy-Based Models with Diffusion Contrastive Divergences
In image generation experiments, the proposed DCD is capable of training an energy-based model for generating the Celab-A $32\times 32$ dataset, which is comparable to existing EBMs.
Entropy-based Training Methods for Scalable Neural Implicit Sampler
These benchmarks include sampling from 2D targets, Bayesian inference, and sampling from high-dimensional energy-based models (EBMs).
Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Models
To demonstrate the effectiveness and universality of Diff-Instruct, we consider two scenarios: distilling pre-trained diffusion models and refining existing GAN models.
End-to-end Training and Decoding for Pivot-based Cascaded Translation Model
Utilizing pivot language effectively can significantly improve low-resource machine translation.
Evaluating the Efficacy of Length-Controllable Machine Translation
We can use automatic summarization or machine translation evaluation metrics for length-controllable machine translation, but this is not necessarily suitable and accurate.
Semi-Infinitely Constrained Markov Decision Processes and Efficient Reinforcement Learning
To the best of our knowledge, we are the first to apply tools from semi-infinitely programming (SIP) to solve constrained reinforcement learning problems.
Asymptotic Behaviors and Phase Transitions in Projected Stochastic Approximation: A Jump Diffusion Approach
Additionally, we propose the Debiased LPSA (DLPSA) as a practical application of our jump diffusion approximation result.
Online Statistical Inference for Nonlinear Stochastic Approximation with Markovian Data
We study the statistical inference of nonlinear stochastic approximation algorithms utilizing a single trajectory of Markovian data.
Robust Markov Decision Processes without Model Estimation
Moreover, we prove the alternative form still plays a similar role as the original form.
Statistical Estimation of Confounded Linear MDPs: An Instrumental Variable Approach
Specifically, we propose a two-stage estimator based on the instrumental variables and establish its statistical properties in the confounded MDPs with a linear structure.
Sparse Adversarial Attack in Multi-agent Reinforcement Learning
Cooperative multi-agent reinforcement learning (cMARL) has many real applications, but the policy trained by existing cMARL algorithms is not robust enough when deployed.
On the Convergence of Policy in Unregularized Policy Mirror Descent
In this short note, we give the convergence analysis of the policy in the recent famous policy mirror descent (PMD).
Federated Reinforcement Learning with Environment Heterogeneity
We study a Federated Reinforcement Learning (FedRL) problem in which $n$ agents collaboratively learn a single policy without sharing the trajectories they collected during agent-environment interaction.
Global Convergence Analysis of Deep Linear Networks with A One-neuron Layer
In this paper, we follow Eftekhari's work to give a non-local convergence analysis of deep linear networks.
A Statistical Analysis of Polyak-Ruppert Averaged Q-learning
We study Q-learning with Polyak-Ruppert averaging in a discounted Markov decision process in synchronous and tabular settings.
Faster Directional Convergence of Linear Neural Networks under Spherically Symmetric Data
In this paper, we study gradient methods for training deep linear neural networks with binary cross-entropy loss.
Greedy and Random Quasi-Newton Methods with Faster Explicit Superlinear Convergence
In this paper, we follow Rodomanov and Nesterov’s work to study quasi-Newton methods.
Statistical Estimation and Inference via Local SGD in Federated Learning
Both the methods are communication efficient and applicable to online data.
Near Optimal Stochastic Algorithms for Finite-Sum Unbalanced Convex-Concave Minimax Optimization
This paper considers stochastic first-order algorithms for convex-concave minimax problems of the form $\min_{\bf x}\max_{\bf y}f(\bf x, \bf y)$, where $f$ can be presented by the average of $n$ individual components which are $L$-average smooth.
Memory-Efficient Differentiable Transformer Architecture Search
To this end, we propose a multi-split reversible network and combine it with DARTS.
Towards Theoretical Understandings of Robust Markov Decision Processes: Sample Complexity and Asymptotics
In this paper, we study the non-asymptotic and asymptotic performances of the optimal robust policy and value function of robust Markov Decision Processes(MDPs), where the optimal robust policy and value function are solved only from a generative model.
Directional Convergence Analysis under Spherically Symmetric Distribution
We consider the fundamental problem of learning linear predictors (i. e., separable datasets with zero margin) using neural networks with gradient flow or gradient descent.
Meta-Regularization: An Approach to Adaptive Choice of the Learning Rate in Gradient Descent
We propose \textit{Meta-Regularization}, a novel approach for the adaptive choice of the learning rate in first-order gradient descent methods.
Multi-split Reversible Transformers Can Enhance Neural Machine Translation
Large-scale transformers have been shown the state-of-the-art on neural machine translation.
DIPPA: An improved Method for Bilinear Saddle Point Problems
This paper studies bilinear saddle point problems $\min_{\bf{x}} \max_{\bf{y}} g(\bf{x}) + \bf{x}^{\top} \bf{A} \bf{y} - h(\bf{y})$, where the functions $g, h$ are smooth and strongly-convex.
Lower Complexity Bounds of Finite-Sum Optimization Problems: The Results and Construction
This construction is friendly to the analysis of PIFO algorithms.
FedPower: Privacy-Preserving Distributed Eigenspace Estimation
The low communication power and the possible privacy breaches of data make the computation of eigenspace challenging.
Delayed Projection Techniques for Linearly Constrained Problems: Convergence Rates, Acceleration, and Applications
In this work, we study a novel class of projection-based algorithms for linearly constrained problems (LCPs) which have a lot of applications in statistics, optimization, and machine learning.
On the Landscape of Sparse Linear Networks
Network pruning, or sparse network has a long history and practical significance in modern applications.
Intervention Generative Adversarial Nets
In this paper we propose a novel approach for stabilizing the training process of Generative Adversarial Networks as well as alleviating the mode collapse problem.
Optimal Designs of Gaussian Processes with Budgets for Hyperparameter Optimization
Moreover, comparisons between different initial designs with the same model show the advantage of the proposed optimal design.
Train Once, and Decode As You Like
In this paper we propose a unified approach for supporting different generation manners of machine translation, including autoregressive, semi-autoregressive, and refinement-based non-autoregressive models.
Active Learning Approaches to Enhancing Neural Machine Translation
Active learning is an efficient approach for mitigating data dependency when training neural machine translation (NMT) models.
Finding the Near Optimal Policy via Adaptive Reduced Regularization in MDPs
Regularized MDPs serve as a smooth version of original MDPs.
On the Landscape of One-hidden-layer Sparse Networks and Beyond
We show that linear networks can have no spurious valleys under special sparse structures, and non-linear networks could also admit no spurious valleys under a wide final layer.
Optimal Quantization for Batch Normalization in Neural Network Deployments and Beyond
Quantized Neural Networks (QNNs) use low bit-width fixed-point numbers for representing weight parameters and activations, and are often used in real-world applications due to their saving of computation resources and reproducibility of results.
Intervention Generative Adversarial Networks
In this paper we propose a novel approach for stabilizing the training process of Generative Adversarial Networks as well as alleviating the mode collapse problem.
An Asymptotically Optimal Multi-Armed Bandit Algorithm and Hyperparameter Optimization
The evaluation of hyperparameters, neural architectures, or data augmentation policies becomes a critical model selection problem in advanced deep learning with a large hyperparameter search space.
Communication-Efficient Distributed SVD via Local Power Iterations
As a practical surrogate of OPT, sign-fixing, which uses a diagonal matrix with $\pm 1$ entries as weights, has better computation complexity and stability in experiments.
Fast Generalized Matrix Regression with Applications in Machine Learning
Fast matrix algorithms have become the fundamental tools of machine learning in big data era.
Communication-Efficient Local Decentralized SGD Methods
Recently, the technique of local updates is a powerful tool in centralized settings to improve communication efficiency via periodical communication.
Distillation $\approx$ Early Stopping? Harvesting Dark Knowledge Utilizing Anisotropic Information Retrieval For Overparameterized Neural Network
Assuming that the teacher network is overparameterized, we argue that the teacher network is essentially harvesting dark knowledge from the data via early stopping.
A Novel Analysis Framework of Lower Complexity Bounds for Finite-Sum Optimization
This paper studies the lower bound complexity for the optimization problem whose objective function is the average of $n$ individual smooth convex functions.
Distillation $\approx$ Early Stopping? Harvesting Dark Knowledge Utilizing Anisotropic Information Retrieval For Overparameterized NN
Assuming that the teacher network is overparameterized, we argue that the teacher network is essentially harvesting dark knowledge from the data via early stopping.
A Stochastic Proximal Point Algorithm for Saddle-Point Problems
We consider saddle point problems which objective functions are the average of $n$ strongly convex-concave individual components.
A General Analysis Framework of Lower Complexity Bounds for Finite-Sum Optimization
This paper studies the lower bound complexity for the optimization problem whose objective function is the average of $n$ individual smooth convex functions.
Towards Better Generalization: BP-SVRG in Training Deep Neural Networks
Stochastic variance-reduced gradient (SVRG) is a classical optimization method.
On the Convergence of FedAvg on Non-IID Data
In this paper, we analyze the convergence of \texttt{FedAvg} on non-iid data and establish a convergence rate of $\mathcal{O}(\frac{1}{T})$ for strongly convex and smooth problems, where $T$ is the number of SGDs.
Gram-Gauss-Newton Method: Learning Overparameterized Neural Networks for Regression Problems
First-order methods such as stochastic gradient descent (SGD) are currently the standard algorithm for training deep neural networks.
Distributionally Robust Optimization Leads to Better Generalization: on SGD and Beyond
In this paper, we adopt distributionally robust optimization (DRO) (Ben-Tal et al., 2013) in hope to achieve a better generalization in deep learning tasks.
Hyper-Regularization: An Adaptive Choice for the Learning Rate in Gradient Descent
Specifically, we impose a regularization term on the learning rate via a generalized distance, and cast the joint updating process of the parameter and the learning rate into a maxmin problem.
A Regularized Approach to Sparse Optimal Policy in Reinforcement Learning
We propose and study a general framework for regularized Markov decision processes (MDPs) where the goal is to find an optimal policy that maximizes the expected discounted total reward plus a policy regularization term.
Lipschitz Generative Adversarial Nets
By contrast, Wasserstein GAN (WGAN), where the discriminative function is restricted to 1-Lipschitz, does not suffer from such a gradient uninformativeness problem.
Do Subsampled Newton Methods Work for High-Dimensional Data?
Subsampled Newton methods approximate Hessian matrices through subsampling techniques, alleviating the cost of forming Hessian matrices but using sufficient curvature information.
Accelerated Value Iteration via Anderson Mixing
A2VI is more efficient than the modified policy iteration, which is a classical approximate method for policy evaluation.
Hierarchical Attention: What Really Counts in Various NLP Tasks
Ham achieves a state-of-the-art BLEU score of 0. 26 on Chinese poem generation task and a nearly 6. 5% averaged improvement compared with the existing machine reading comprehension models such as BIDAF and Match-LSTM.
Understanding the Effectiveness of Lipschitz-Continuity in Generative Adversarial Nets
In this paper, we investigate the underlying factor that leads to failure and success in the training of GANs.
Interpolatron: Interpolation or Extrapolation Schemes to Accelerate Optimization for Deep Neural Networks
In this paper we explore acceleration techniques for large scale nonconvex optimization problems with special focuses on deep neural networks.
Nesterov's Acceleration For Approximate Newton
Besides, the accelerated regularized sub-sampled Newton has good performance comparable to or even better than classical algorithms.
Approximate Newton Methods and Their Local Convergence
We propose a unifying framework to analyze local convergence properties of second order methods.
Nestrov's Acceleration For Second Order Method
Besides, the accelerated regularized sub-sampled Newton has good performance comparable to or even better than state-of-art algorithms.
Robust Frequent Directions with Application in Online Learning
We also apply RFD to online learning and propose an effective hyperparameter-free online Newton algorithm.
Communication Lower Bounds for Distributed Convex Optimization: Partition Data on Features
Recently, there has been an increasing interest in designing distributed convex optimization algorithms under the setting where the data matrix is partitioned on features.
An Efficient Character-Level Neural Machine Translation
The encoder-decoder architecture with an attention mechanism achieves a translation performance comparable to the existing state-of-the-art phrase-based systems on the task of English-to-French translation.
A Proximal Stochastic Quasi-Newton Algorithm
It incorporates the Hessian in the smooth part of the function and exploits multistage scheme to reduce the variance of the stochastic gradient.
Wishart Mechanism for Differentially Private Principal Components Analysis
We propose a new input perturbation mechanism for publishing a covariance matrix to achieve $(\epsilon, 0)$-differential privacy.
A New Relaxation Approach to Normalized Hypergraph Cut
Normalized graph cut (NGC) has become a popular research topic due to its wide applications in a large variety of areas like machine learning and very large scale integration (VLSI) circuit design.
The Singular Value Decomposition, Applications and Beyond
Built on SVD and a theory of symmetric gauge functions, we discuss unitarily invariant norms, which are then used to formulate general results for matrix low rank approximation.
Nonconvex Penalization in Sparse Estimation: An Approach Based on the Bernstein Function
In this paper we study nonconvex penalization using Bernstein functions whose first-order derivatives are completely monotone.
A Parallel algorithm for $\mathcal{X}$-Armed bandits
To deal with these large-scale data sets, we study a distributed setting of $\mathcal{X}$-armed bandits, where $m$ players collaborate to find the maximum of the unknown function.
A Scalable and Extensible Framework for Superposition-Structured Models
In many learning tasks, structural models usually lead to better interpretability and higher generalization performance.
Regret vs. Communication: Distributed Stochastic Multi-Armed Bandits and Beyond
When the time horizon is unknown, we measure the frequency of communication through a new notion called the density of the communication set, and give an exact characterization of the interplay between regret and communication.
Towards More Efficient SPSD Matrix Approximation and CUR Matrix Decomposition
The Nystr\"om method is a special instance of our fast model and is approximation to the prototype model.
A Nonconvex Approach for Structured Sparse Learning
In this paper we focus on the $\ell_q$-analysis optimization problem for structured sparse learning ($0< q \leq 1$).
Adjusting Leverage Scores by Row Weighting: A Practical Approach to Coherent Matrix Completion
Low-rank matrix completion is an important problem with extensive real-world applications.
Distributed Power-law Graph Computing: Theoretical and Empirical Analysis
We theoretically prove that DBH can achieve lower communication cost than existing methods and can simultaneously guarantee good workload balance.
Group Orbit Optimization: A Unified Approach to Data Normalization
In this paper we propose and study an optimization problem over a matrix group orbit that we call \emph{Group Orbit Optimization} (GOO).
SPSD Matrix Approximation vis Column Selection: Theories, Algorithms, and Extensions
In this paper we conduct in-depth studies of an SPSD matrix approximation model and establish strong relative-error bounds.
Efficient Algorithms and Error Analysis for the Modified Nystrom Method
Recently, a variant of the Nystr\"om method called the modified Nystr\"om method has demonstrated significant improvement over the standard Nystr\"om method in approximation accuracy, both theoretically and empirically.
The Bernstein Function: A Unifying Framework of Nonconvex Penalization in Sparse Estimation
In this paper we study nonconvex penalization using Bernstein functions.
The Matrix Ridge Approximation: Algorithms and Applications
We are concerned with an approximation problem for a symmetric positive semidefinite matrix due to motivation from a class of nonlinear machine learning methods.
Compound Poisson Processes, Latent Shrinkage Priors and Bayesian Nonconvex Penalization
In this paper we discuss Bayesian nonconvex penalization for sparse learning problems.
Kinetic Energy Plus Penalty Functions for Sparse Estimation
In this paper we propose and study a family of sparsity-inducing penalty functions.
Improving CUR Matrix Decomposition and the Nyström Approximation via Adaptive Sampling
The CUR matrix decomposition and the Nystr\"{o}m approximation are two important low-rank matrix approximation techniques.
A Scalable CUR Matrix Decomposition Algorithm: Lower Time Complexity and Tighter Bound
The CUR matrix decomposition is an important extension of Nyström approximation to a general matrix.
Optimal Scoring for Unsupervised Learning
We are often interested in casting classification and clustering problems in a regression framework, because it is feasible to achieve some statistical properties in this framework by imposing some penalty criteria.
Posterior Consistency of the Silverman g-prior in Bayesian Model Choice
The duality between regularization and prior leads to interpreting regularization methods in terms of maximum a posteriori estimation and has motivated Bayesian interpretations of kernel methods.
