Search Results for author:
Found 82 papers, 17 papers with code
Steering No-Regret Agents in MFGs under Model Uncertainty
Our work presents an effective framework for steering agents behaviors in large-population systems under uncertainty.
Can RLHF be More Efficient with Imperfect Reward Models? A Policy Coverage Perspective
This paper studies how to transfer knowledge from those imperfect reward models in online RLHF.
Poincaré Inequality for Local Log-Polyak-Lojasiewicz Measures : Non-asymptotic Analysis in Low-temperature Regime
To understand the convergence behavior of stochastic dynamics in such landscapes, we propose to study the class of \logPLmeasure\ measures $\mu_\epsilon \propto \exp(-V/\epsilon)$, where the potential $V$ satisfies a local Polyak-{\L}ojasiewicz (P\L) inequality, and its set of local minima is provably \emph{connected}.
On the Crucial Role of Initialization for Matrix Factorization
This work revisits the classical low-rank matrix factorization problem and unveils the critical role of initialization in shaping convergence rates for such nonconvex and nonsmooth optimization.
Implicit Regularization of Sharpness-Aware Minimization for Scale-Invariant Problems
Sharpness-aware minimization (SAM) improves generalization of various deep learning tasks.
From Gradient Clipping to Normalization for Heavy Tailed SGD
In the setting where all problem parameters are known, we show this complexity is improved to $\mathcal{O}\left(\varepsilon^{-\frac{3p-2}{p-1}}\right)$, matching the previously known lower bound for all first-order methods in all problem dependent parameters.
Exploiting Approximate Symmetry for Efficient Multi-Agent Reinforcement Learning
We show that TD learning converges up to a small bias using trajectories of the $N$-player game with finite-sample guarantees, permitting symmetrized learning without building an explicit MFG model.
Multi-agent Reinforcement Learning
reinforcement-learning
+1
Multi-level Monte-Carlo Gradient Methods for Stochastic Optimization with Biased Oracles
This setting captures various optimization paradigms, such as conditional stochastic optimization, distributionally robust optimization, shortfall risk optimization, and machine learning paradigms, such as contrastive learning.
Independent Policy Mirror Descent for Markov Potential Games: Scaling to Large Number of Players
In this work, we investigate the iteration complexity of an independent policy mirror descent (PMD) algorithm for MPGs.
Complexity of Minimizing Projected-Gradient-Dominated Functions with Stochastic First-order Oracles
Furthermore, we show that a projected variance-reduced first-order algorithm can obtain the upper complexity bound of $\mathcal{O}(\epsilon^{-{2}/{\alpha}})$, matching the lower bound.
Learning to Steer Markovian Agents under Model Uncertainty
We introduce a model-based non-episodic Reinforcement Learning (RL) formulation for our steering problem.
Achieving Near-Optimal Convergence for Distributed Minimax Optimization with Adaptive Stepsizes
This protocol ensures the consistency among stepsizes of nodes, eliminating the steady-state error due to the lack of coordination of stepsizes among nodes that commonly exists in vanilla distributed adaptive methods, and thus guarantees exact convergence.
A Hessian-Aware Stochastic Differential Equation for Modelling SGD
Continuous-time approximation of Stochastic Gradient Descent (SGD) is a crucial tool to study its escaping behaviors from stationary points.
Primal Methods for Variational Inequality Problems with Functional Constraints
These algorithms along with their theoretical analysis often require the existence and prior knowledge of the optimal Lagrange multipliers.
Independent Learning in Constrained Markov Potential Games
We propose an independent policy gradient algorithm for learning approximate constrained Nash equilibria: Each agent observes their own actions and rewards, along with a shared state.
Taming Nonconvex Stochastic Mirror Descent with General Bregman Divergence
This paper revisits the convergence of Stochastic Mirror Descent (SMD) in the contemporary nonconvex optimization setting.
Truly No-Regret Learning in Constrained MDPs
As Efroni et al. (2020) pointed out, it is an open question whether primal-dual algorithms can provably achieve sublinear regret if we do not allow error cancellations.
Model-Based RL for Mean-Field Games is not Statistically Harder than Single-Agent RL
We study the sample complexity of reinforcement learning (RL) in Mean-Field Games (MFGs) with model-based function approximation that requires strategic exploration to find a Nash Equilibrium policy.
Efficiently Escaping Saddle Points for Non-Convex Policy Optimization
Policy gradient (PG) is widely used in reinforcement learning due to its scalability and good performance.
Parameter-Agnostic Optimization under Relaxed Smoothness
However, as the assumption is relaxed to the more realistic $(L_0, L_1)$-smoothness, all existing convergence results still necessitate tuning of the stepsize.
Optimal Guarantees for Algorithmic Reproducibility and Gradient Complexity in Convex Optimization
Particularly, given the inexact initialization oracle, our regularization-based algorithms achieve the best of both worlds - optimal reproducibility and near-optimal gradient complexity - for minimization and minimax optimization.
DPZero: Private Fine-Tuning of Language Models without Backpropagation
The widespread practice of fine-tuning large language models (LLMs) on domain-specific data faces two major challenges in memory and privacy.
A Convex Framework for Confounding Robust Inference
We study policy evaluation of offline contextual bandits subject to unobserved confounders.
Learning Zero-Sum Linear Quadratic Games with Improved Sample Complexity and Last-Iterate Convergence
Our main results are two-fold: (i) in the deterministic setting, we establish the first global last-iterate linear convergence result for the nested algorithm that seeks NE of zero-sum LQ games; (ii) in the model-free setting, we establish a~$\widetilde{\mathcal{O}}(\epsilon^{-2})$ sample complexity using a single-point ZO estimator.
Provably Convergent Policy Optimization via Metric-aware Trust Region Methods
Trust-region methods based on Kullback-Leibler divergence are pervasively used to stabilize policy optimization in reinforcement learning.
Provably Learning Nash Policies in Constrained Markov Potential Games
Multi-agent reinforcement learning (MARL) addresses sequential decision-making problems with multiple agents, where each agent optimizes its own objective.
Cancellation-Free Regret Bounds for Lagrangian Approaches in Constrained Markov Decision Processes
Unlike existing Lagrangian approaches, our algorithm achieves this regret without the need for the cancellation of errors.
Reinforcement Learning with General Utilities: Simpler Variance Reduction and Large State-Action Space
We consider the reinforcement learning (RL) problem with general utilities which consists in maximizing a function of the state-action occupancy measure.
On the Statistical Efficiency of Mean-Field Reinforcement Learning with General Function Approximation
In this paper, we study the fundamental statistical efficiency of Reinforcement Learning in Mean-Field Control (MFC) and Mean-Field Game (MFG) with general model-based function approximation.
Kernel Conditional Moment Constraints for Confounding Robust Inference
It can be shown that our estimator contains the recently proposed sharp estimator by Dorn and Guo (2022) as a special case, and our method enables a novel extension of the classical marginal sensitivity model using f-divergence.
Robust Knowledge Transfer in Tiered Reinforcement Learning
In this paper, we study the Tiered Reinforcement Learning setting, a parallel transfer learning framework, where the goal is to transfer knowledge from the low-tier (source) task to the high-tier (target) task to reduce the exploration risk of the latter while solving the two tasks in parallel.
Stochastic Policy Gradient Methods: Improved Sample Complexity for Fisher-non-degenerate Policies
Recently, the impressive empirical success of policy gradient (PG) methods has catalyzed the development of their theoretical foundations.
Policy Mirror Ascent for Efficient and Independent Learning in Mean Field Games
Instead, we show that $N$ agents running policy mirror ascent converge to the Nash equilibrium of the regularized game within $\widetilde{\mathcal{O}}(\varepsilon^{-2})$ samples from a single sample trajectory without a population generative model, up to a standard $\mathcal{O}(\frac{1}{\sqrt{N}})$ error due to the mean field.
Learning to Optimize with Stochastic Dominance Constraints
In real-world decision-making, uncertainty is important yet difficult to handle.
TiAda: A Time-scale Adaptive Algorithm for Nonconvex Minimax Optimization
Adaptive gradient methods have shown their ability to adjust the stepsizes on the fly in a parameter-agnostic manner, and empirically achieve faster convergence for solving minimization problems.
Finite-Time Analysis of Entropy-Regularized Neural Natural Actor-Critic Algorithm
Natural actor-critic (NAC) and its variants, equipped with the representation power of neural networks, have demonstrated impressive empirical success in solving Markov decision problems with large state spaces.
Nest Your Adaptive Algorithm for Parameter-Agnostic Nonconvex Minimax Optimization
Adaptive algorithms like AdaGrad and AMSGrad are successful in nonconvex optimization owing to their parameter-agnostic ability -- requiring no a priori knowledge about problem-specific parameters nor tuning of learning rates.
Bring Your Own Algorithm for Optimal Differentially Private Stochastic Minimax Optimization
We provide a general framework for solving differentially private stochastic minimax optimization (DP-SMO) problems, which enables the practitioners to bring their own base optimization algorithm and use it as a black-box to obtain the near-optimal privacy-loss trade-off.
Generalization Bounds of Nonconvex-(Strongly)-Concave Stochastic Minimax Optimization
We further study the algorithm-dependent generalization bounds via stability arguments of algorithms.
Stochastic Second-Order Methods Improve Best-Known Sample Complexity of SGD for Gradient-Dominated Function
We prove that the total sample complexity of SCRN in achieving $\epsilon$-global optimum is $\mathcal{O}(\epsilon^{-7/(2\alpha)+1})$ for $1\le\alpha< 3/2$ and $\mathcal{\tilde{O}}(\epsilon^{-2/(\alpha)})$ for $3/2\le\alpha\le 2$.
Momentum-Based Policy Gradient with Second-Order Information
SHARP algorithm is parameter-free, achieving $\epsilon$-approximate first-order stationary point with $O(\epsilon^{-3})$ number of trajectories, while using a batch size of $O(1)$ at each iteration.
Finite-Time Analysis of Natural Actor-Critic for POMDPs
We consider the reinforcement learning problem for partially observed Markov decision processes (POMDPs) with large or even countably infinite state spaces, where the controller has access to only noisy observations of the underlying controlled Markov chain.
Lifted Primal-Dual Method for Bilinearly Coupled Smooth Minimax Optimization
We also provide a direct single-loop algorithm, using the LPD method, that achieves the iteration complexity of $O(\sqrt{\frac{L_x}{\varepsilon}} + \frac{\|A\|}{\sqrt{\mu_y \varepsilon}} + \sqrt{\frac{L_y}{\varepsilon}})$.
Faster Single-loop Algorithms for Minimax Optimization without Strong Concavity
Gradient descent ascent (GDA), the simplest single-loop algorithm for nonconvex minimax optimization, is widely used in practical applications such as generative adversarial networks (GANs) and adversarial training.
On the Bias-Variance-Cost Tradeoff of Stochastic Optimization
We consider stochastic optimization when one only has access to biased stochastic oracles of the objective, and obtaining stochastic gradients with low biases comes at high costs.
Sample-efficient actor-critic algorithms with an etiquette for zero-sum Markov games
Our sample complexities also match the best-known results for global convergence of policy gradient and two time-scale actor-critic algorithms in the single agent setting.
Efficient Wasserstein and Sinkhorn Policy Optimization
Trust-region methods based on Kullback-Leibler divergence are pervasively used to stabilize policy optimization in reinforcement learning.
Linear Convergence of Entropy-Regularized Natural Policy Gradient with Linear Function Approximation
Furthermore, under mild regularity conditions on the concentrability coefficient and basis vectors, we prove that entropy-regularized NPG exhibits \emph{linear convergence} up to a function approximation error.
The Complexity of Nonconvex-Strongly-Concave Minimax Optimization
In the averaged smooth finite-sum setting, our proposed algorithm improves over previous algorithms by providing a nearly-tight dependence on the condition number.
Simulation Studies on Deep Reinforcement Learning for Building Control with Human Interaction
The building sector consumes the largest energy in the world, and there have been considerable research interests in energy consumption and comfort management of buildings.
Sample Complexity and Overparameterization Bounds for Temporal Difference Learning with Neural Network Approximation
In this paper, we study the dynamics of temporal difference learning with neural network-based value function approximation over a general state space, namely, \emph{Neural TD learning}.
A Discrete-Time Switching System Analysis of Q-learning
Based on these two systems, we derive a new finite-time error bound of asynchronous Q-learning when a constant stepsize is used.
A Unified Switching System Perspective and Convergence Analysis of Q-Learning Algorithms
This paper develops a novel and unified framework to analyze the convergence of a large family of Q-learning algorithms from the switching system perspective.
The Devil is in the Detail: A Framework for Macroscopic Prediction via Microscopic Models
Macroscopic data aggregated from microscopic events are pervasive in machine learning, such as country-level COVID-19 infection statistics based on city-level data.
A Catalyst Framework for Minimax Optimization
We introduce a generic \emph{two-loop} scheme for smooth minimax optimization with strongly-convex-concave objectives.
Global Convergence and Variance Reduction for a Class of Nonconvex-Nonconcave Minimax Problems
Nonconvex minimax problems appear frequently in emerging machine learning applications, such as generative adversarial networks and adversarial learning.
The Mean-Squared Error of Double Q-Learning
In this paper, we establish a theoretical comparison between the asymptotic mean-squared error of Double Q-learning and Q-learning.
Biased Stochastic First-Order Methods for Conditional Stochastic Optimization and Applications in Meta Learning
Conditional stochastic optimization covers a variety of applications ranging from invariant learning and causal inference to meta-learning.
Periodic Q-Learning
The use of target networks is a common practice in deep reinforcement learning for stabilizing the training; however, theoretical understanding of this technique is still limited.
Global Convergence and Variance-Reduced Optimization for a Class of Nonconvex-Nonconcave Minimax Problems
Nonconvex minimax problems appear frequently in emerging machine learning applications, such as generative adversarial networks and adversarial learning.
A Unified Switching System Perspective and O.D.E. Analysis of Q-Learning Algorithms
In this paper, we introduce a unified framework for analyzing a large family of Q-learning algorithms, based on switching system perspectives and ODE-based stochastic approximation.
Learning Positive Functions with Pseudo Mirror Descent
The nonparametric learning of positive-valued functions appears widely in machine learning, especially in the context of estimating intensity functions of point processes.
Optimization for Reinforcement Learning: From Single Agent to Cooperative Agents
This article reviews recent advances in multi-agent reinforcement learning algorithms for large-scale control systems and communication networks, which learn to communicate and cooperate.
Distributed Optimization
Multi-agent Reinforcement Learning
+3
Sample Complexity of Sample Average Approximation for Conditional Stochastic Optimization
In this paper, we study a class of stochastic optimization problems, referred to as the \emph{Conditional Stochastic Optimization} (CSO), in the form of $\min_{x \in \mathcal{X}} \EE_{\xi}f_\xi\Big({\EE_{\eta|\xi}[g_\eta(x,\xi)]}\Big)$, which finds a wide spectrum of applications including portfolio selection, reinforcement learning, robust learning, causal inference and so on.
Exponential Family Estimation via Adversarial Dynamics Embedding
We present an efficient algorithm for maximum likelihood estimation (MLE) of exponential family models, with a general parametrization of the energy function that includes neural networks.
Target-Based Temporal Difference Learning
The use of target networks has been a popular and key component of recent deep Q-learning algorithms for reinforcement learning, yet little is known from the theory side.
Quadratic Decomposable Submodular Function Minimization: Theory and Practice (Computation and Analysis of PageRank over Hypergraphs)
We introduce a new convex optimization problem, termed quadratic decomposable submodular function minimization (QDSFM), which allows to model a number of learning tasks on graphs and hypergraphs.
Coupled Variational Bayes via Optimization Embedding
This flexible function class couples the variational distribution with the original parameters in the graphical models, allowing end-to-end learning of the graphical models by back-propagation through the variational distribution.
Predictive Approximate Bayesian Computation via Saddle Points
Approximate Bayesian computation (ABC) is an important methodology for Bayesian inference when the likelihood function is intractable.
Kernel Exponential Family Estimation via Doubly Dual Embedding
We investigate penalized maximum log-likelihood estimation for exponential family distributions whose natural parameter resides in a reproducing kernel Hilbert space.
Quadratic Decomposable Submodular Function Minimization
The problem is closely related to decomposable submodular function minimization and arises in many learning on graphs and hypergraphs settings, such as graph-based semi-supervised learning and PageRank.
Nonparametric Hawkes Processes: Online Estimation and Generalization Bounds
In this paper, we design a nonparametric online algorithm for estimating the triggering functions of multivariate Hawkes processes.
SBEED: Convergent Reinforcement Learning with Nonlinear Function Approximation
When function approximation is used, solving the Bellman optimality equation with stability guarantees has remained a major open problem in reinforcement learning for decades.
Boosting the Actor with Dual Critic
This paper proposes a new actor-critic-style algorithm called Dual Actor-Critic or Dual-AC.
Online Learning for Multivariate Hawkes Processes
We develop a nonparametric and online learning algorithm that estimates the triggering functions of a multivariate Hawkes process (MHP).
Stochastic Generative Hashing
Learning-based binary hashing has become a powerful paradigm for fast search and retrieval in massive databases.
Fast and Simple Optimization for Poisson Likelihood Models
Since almost all gradient-based optimization algorithms rely on Lipschitz-continuity, optimizing Poisson likelihood models with a guarantee of convergence can be challenging, especially for large-scale problems.
Learning from Conditional Distributions via Dual Embeddings
In such problems, each sample $x$ itself is associated with a conditional distribution $p(z|x)$ represented by samples $\{z_i\}_{i=1}^M$, and the goal is to learn a function $f$ that links these conditional distributions to target values $y$.
Time-Sensitive Recommendation From Recurrent User Activities
By making personalized suggestions, a recommender system is playing a crucial role in improving the engagement of users in modern web-services.
Semi-proximal Mirror-Prox for Nonsmooth Composite Minimization
We propose a new first-order optimisation algorithm to solve high-dimensional non-smooth composite minimisation problems.
Provable Bayesian Inference via Particle Mirror Descent
Bayesian methods are appealing in their flexibility in modeling complex data and ability in capturing uncertainty in parameters.
Scalable Kernel Methods via Doubly Stochastic Gradients
The general perception is that kernel methods are not scalable, and neural nets are the methods of choice for nonlinear learning problems.
