English
Related papers

Related papers: Improved Sample Complexity Analysis of Natural Pol…

200 papers

Novel advanced policy gradient (APG) methods, such as Trust Region policy optimization and Proximal policy optimization (PPO), have become the dominant reinforcement learning algorithms because of their ease of implementation and good…

Optimization and Control · Mathematics 2022-03-22 J. G. Dai , Mark Gluzman

We study the convergence rate of first-order methods for rectangular matrix factorization, which is a canonical nonconvex optimization problem. Specifically, given a rank-$r$ matrix $\mathbf{A}\in\mathbb{R}^{m\times n}$, we prove that…

Machine Learning · Computer Science 2024-12-03 Zhenghao Xu , Yuqing Wang , Tuo Zhao , Rachel Ward , Molei Tao

Markov Decision Processes (MDPs) have been used to formulate many decision-making problems in science and engineering. The objective is to synthesize the best decision (action selection) policies to maximize expected rewards (minimize…

Optimization and Control · Mathematics 2015-07-08 Mahmoud El Chamie , Behcet Acikmese

We study the problem of minimizing a strongly convex, smooth function when we have noisy estimates of its gradient. We propose a novel multistage accelerated algorithm that is universally optimal in the sense that it achieves the optimal…

Optimization and Control · Mathematics 2019-10-29 Necdet Serhat Aybat , Alireza Fallah , Mert Gurbuzbalaban , Asuman Ozdaglar

We propose a new policy gradient method, named homotopic policy mirror descent (HPMD), for solving discounted, infinite horizon MDPs with finite state and action spaces. HPMD performs a mirror descent type policy update with an additional…

Machine Learning · Computer Science 2022-11-30 Yan Li , Guanghui Lan , Tuo Zhao

Partially Observable Markov Decision Processes (POMDPs) are powerful models for sequential decision making under transition and observation uncertainties. This paper studies the challenging yet important problem in POMDPs known as the…

Artificial Intelligence · Computer Science 2024-06-06 Qi Heng Ho , Martin S. Feather , Federico Rossi , Zachary N. Sunberg , Morteza Lahijanian

We study best-policy identification for finite-horizon risk-sensitive reinforcement learning under the entropic risk measure. Recent work established a constant gap in the exponential horizon dependence between lower and upper bounds on the…

Machine Learning · Computer Science 2026-05-14 Amer Essakine , Claire Vernade

The policy gradient theorem is defined based on an objective with respect to the initial distribution over states. In the discounted case, this results in policies that are optimal for one distribution over initial states, but may not be…

Machine Learning · Computer Science 2019-12-12 Riashat Islam , Raihan Seraj , Pierre-Luc Bacon , Doina Precup

We analyze the convergence rate of the unregularized natural policy gradient algorithm with log-linear policy parametrizations in infinite-horizon discounted Markov decision processes. In the deterministic case, when the Q-value is known…

Machine Learning · Computer Science 2023-03-15 Carlo Alfano , Patrick Rebeschini

Consider the stochastic composition optimization problem where the objective is a composition of two expected-value functions. We propose a new stochastic first-order method, namely the accelerated stochastic compositional proximal gradient…

Optimization and Control · Mathematics 2016-07-26 Mengdi Wang , Ji Liu , Ethan X. Fang

Control design for robotic systems is complex and often requires solving an optimization to follow a trajectory accurately. Online optimization approaches like Model Predictive Control (MPC) have been shown to achieve great tracking…

We address the problem of finding the optimal policy of a constrained Markov decision process (CMDP) using a gradient descent-based algorithm. Previous results have shown that a primal-dual approach can achieve an $\mathcal{O}(1/\sqrt{T})$…

Machine Learning · Computer Science 2022-02-07 Tao Liu , Ruida Zhou , Dileep Kalathil , P. R. Kumar , Chao Tian

We consider infinite-horizon stationary $\gamma$-discounted Markov Decision Processes, for which it is known that there exists a stationary optimal policy. Using Value and Policy Iteration with some error $\epsilon$ at each iteration, it is…

Machine Learning · Computer Science 2012-11-30 Bruno Scherrer , Boris Lesner

Large-scale non-convex sparsity-constrained problems have recently gained extensive attention. Most existing deterministic optimization methods (e.g., GraSP) are not suitable for large-scale and high-dimensional problems, and thus…

Machine Learning · Computer Science 2019-12-03 Fanhua Shang , Bingkun Wei , Hongying Liu , Yuanyuan Liu , Jiacheng Zhuo

This paper investigates the optimization problem of an infinite stage discrete time Markov decision process (MDP) with a long-run average metric considering both mean and variance of rewards together. Such performance metric is important…

Optimization and Control · Mathematics 2020-08-11 Li Xia

We investigate the classical active pure exploration problem in Markov Decision Processes, where the agent sequentially selects actions and, from the resulting system trajectory, aims at identifying the best policy as fast as possible. We…

Machine Learning · Statistics 2021-10-26 Aymen Al Marjani , Aurélien Garivier , Alexandre Proutiere

We are interested in risk constraints for infinite horizon discrete time Markov decision processes (MDPs). Starting with average reward MDPs, we show that increasing concave stochastic dominance constraints on the empirical distribution of…

Optimization and Control · Mathematics 2012-06-21 William B. Haskell , Rahul Jain

Policy gradient (PG) methods have played an essential role in the empirical successes of reinforcement learning. In order to handle large state-action spaces, PG methods are typically used with function approximation. In this setting, the…

Policy gradient methods are extensively used in reinforcement learning as a way to optimize expected return. In this paper, we explore the evolution of the policy parameters, for a special class of exactly solvable POMDPs, as a…

Machine Learning · Computer Science 2020-11-04 Gavin McCracken , Colin Daniels , Rosie Zhao , Anna Brandenberger , Prakash Panangaden , Doina Precup

The policy gradient approach is a flexible and powerful reinforcement learning method particularly for problems with continuous actions such as robot control. A common challenge in this scenario is how to reduce the variance of policy…

Machine Learning · Computer Science 2013-01-18 Tingting Zhao , Hirotaka Hachiya , Voot Tangkaratt , Jun Morimoto , Masashi Sugiyama