English
Related papers

Related papers: Variance Reduction for Policy Gradient with Action…

200 papers

When designing algorithms for finite-time-horizon episodic reinforcement learning problems, a common approach is to introduce a fictitious discount factor and use stationary policies for approximations. Empirically, it has been shown that…

Machine Learning · Computer Science 2021-09-15 Xin Guo , Anran Hu , Junzi Zhang

The empirical success of multi-agent reinforcement learning (MARL) has motivated the search for more efficient and scalable algorithms for large scale multi-agent systems. However, existing state-of-the-art algorithms do not fully exploit…

Multiagent Systems · Computer Science 2025-10-14 Shahbaz P Qadri Syed , He Bai

The task of estimating the gradient of a function in the presence of noise is central to several forms of reinforcement learning, including policy search methods. We present two techniques for reducing gradient estimation errors in the…

Machine Learning · Computer Science 2012-12-12 Gregory Lawrence , Noah Cowan , Stuart Russell

In reinforcement learning (RL), offline learning decoupled learning from data collection and is useful in dealing with exploration-exploitation tradeoff and enables data reuse in many applications. In this work, we study two offline…

Machine Learning · Computer Science 2022-02-08 Jing Dong , Xin T. Tong

Learning task-oriented dialog policies via reinforcement learning typically requires large amounts of interaction with users, which in practice renders such methods unusable for real-world applications. In order to reduce the data…

Computation and Language · Computer Science 2022-07-04 Jorge A. Mendez , Alborz Geramifard , Mohammad Ghavamzadeh , Bing Liu

Due to the high variance of policy gradients, on-policy optimization algorithms are plagued with low sample efficiency. In this work, we propose Augment-Reinforce-Merge (ARM) policy gradient estimator as an unbiased low-variance alternative…

Machine Learning · Computer Science 2019-03-14 Yunhao Tang , Mingzhang Yin , Mingyuan Zhou

This paper studies an infinite horizon optimal control problem for discrete-time linear system and quadratic criteria, both with random parameters which are independent and identically distributed with respect to time. In this general…

Optimization and Control · Mathematics 2024-03-04 Deyue Li

While deep reinforcement learning has achieved tremendous successes in various applications, most existing works only focus on maximizing the expected value of total return and thus ignore its inherent stochasticity. Such stochasticity is…

Machine Learning · Computer Science 2023-09-19 Han Zhong , Xun Deng , Ethan X. Fang , Zhuoran Yang , Zhaoran Wang , Runze Li

We provide performance guarantees for a variant of simulation-based policy iteration for controlling Markov decision processes that involves the use of stochastic approximation algorithms along with state-of-the-art techniques that are…

Machine Learning · Computer Science 2022-10-17 Anna Winnicki , R. Srikant

Sample inefficiency is a long-lasting problem in reinforcement learning (RL). The state-of-the-art estimates the optimal action values while it usually involves an extensive search over the state-action space and unstable optimization.…

Machine Learning · Computer Science 2019-11-27 Kaixiang Lin , Jiayu Zhou

We present the first finite time global convergence analysis of policy gradient in the context of infinite horizon average reward Markov decision processes (MDPs). Specifically, we focus on ergodic tabular MDPs with finite state and action…

Machine Learning · Computer Science 2024-03-12 Navdeep Kumar , Yashaswini Murthy , Itai Shufaro , Kfir Y. Levy , R. Srikant , Shie Mannor

Transferring reinforcement learning policies trained in physics simulation to the real hardware remains a challenge, known as the "sim-to-real" gap. Domain randomization is a simple yet effective technique to address dynamics discrepancies…

Robotics · Computer Science 2021-04-05 Ioannis Exarchos , Yifeng Jiang , Wenhao Yu , C. Karen Liu

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a powerful paradigm for post-training large reasoning models (LRMs) using policy-gradient methods such as GRPO. To stabilize training, these methods typically center…

Machine Learning · Computer Science 2026-02-19 Guanning Zeng , Zhaoyi Zhou , Daman Arora , Andrea Zanette

We study variance-dependent regret bounds for Markov decision processes (MDPs). Algorithms with variance-dependent regret guarantees can automatically exploit environments with low variance (e.g., enjoying constant regret on deterministic…

Machine Learning · Computer Science 2023-05-23 Runlong Zhou , Zihan Zhang , Simon S. Du

Humans are masters at quickly learning many complex tasks, relying on an approximate understanding of the dynamics of their environments. In much the same way, we would like our learning agents to quickly adapt to new tasks. In this paper,…

In probably approximately correct (PAC) reinforcement learning (RL), an agent is required to identify an $\epsilon$-optimal policy with probability $1-\delta$. While minimax optimal algorithms exist for this problem, its instance-dependent…

Machine Learning · Computer Science 2022-10-25 Andrea Tirinzoni , Aymen Al-Marjani , Emilie Kaufmann

Policy gradient methods are notorious for having a large variance and high sample complexity. To mitigate this, we introduce SoftTreeMax -- a generalization of softmax that employs planning. In SoftTreeMax, we extend the traditional logits…

Machine Learning · Computer Science 2025-05-27 Gal Dalal , Assaf Hallak , Gugan Thoppe , Shie Mannor , Gal Chechik

The beneficial effects of treatments vary across individuals in most studies. Treatment heterogeneity motivates practitioners to search for the optimal policy based on personal characteristics. A long-standing common practice in policy…

Statistics Theory · Mathematics 2025-01-06 Xuqiao Li , Ying Yan

We revisit the estimation bias in policy gradients for the discounted episodic Markov decision process (MDP) from Deep Reinforcement Learning (DRL) perspective. The objective is formulated theoretically as the expected returns discounted…

Machine Learning · Computer Science 2023-02-13 Haoxuan Pan , Deheng Ye , Xiaoming Duan , Qiang Fu , Wei Yang , Jianping He , Mingfei Sun

An ongoing challenge in neural information processing is: how do neurons adjust their connectivity to improve task performance over time (i.e., actualize learning)? It is widely believed that there is a consistent, synaptic-level learning…

Neural and Evolutionary Computing · Computer Science 2021-06-01 Aman Bhargava , Mohammad R. Rezaei , Milad Lankarany