English
Related papers

Related papers: Doubly Robust Off-Policy Actor-Critic: Convergence…

200 papers

In this paper, we establish last-iterate convergence rates for off-policy actor--critic methods in reinforcement learning. In particular, under a single-loop, single-timescale implementation and a broad class of policy updates, including…

Machine Learning · Computer Science 2026-05-14 Ishaq Hamza , Zaiwei Chen

In Volt/Var control (VVC) of active distribution networks(ADNs), both slow timescale discrete devices (STDDs) and fast timescale continuous devices (FTCDs) are involved. The STDDs such as on-load tap changers (OLTC) and FTCDs such as…

Systems and Control · Electrical Eng. & Systems 2021-04-14 Haotian Liu , Wenchuan Wu

The average reward criterion is relatively less studied as most existing works in the Reinforcement Learning literature consider the discounted reward criterion. There are few recent works that present on-policy average reward actor-critic…

Machine Learning · Computer Science 2023-07-20 Naman Saxena , Subhojyoti Khastigir , Shishir Kolathaya , Shalabh Bhatnagar

This work extends an established critic match loss landscape visualization method from online to off-policy reinforcement learning (RL), aiming to reveal the optimization geometry behind critic learning. Off-policy RL differs from stepwise…

Machine Learning · Computer Science 2026-03-17 Jingyi Liu , Jian Guo , Eberhard Gill

Value-based reinforcement-learning algorithms provide state-of-the-art results in model-free discrete-action settings, and tend to outperform actor-critic algorithms. We argue that actor-critic algorithms are limited by their need for an…

Machine Learning · Computer Science 2019-06-13 Denis Steckelmacher , Hélène Plisnier , Diederik M. Roijers , Ann Nowé

In this work, we consider the off-policy policy evaluation problem for contextual bandits and finite horizon reinforcement learning in the nonstationary setting. Reusing old data is critical for policy evaluation, but existing estimators…

Machine Learning · Computer Science 2023-02-24 Vincent Liu , Yash Chandak , Philip Thomas , Martha White

Actor-critic (AC) is a powerful method for learning an optimal policy in reinforcement learning, where the critic uses algorithms, e.g., temporal difference (TD) learning with function approximation, to evaluate the current policy and the…

Machine Learning · Computer Science 2024-06-05 Yudan Wang , Yue Wang , Yi Zhou , Shaofeng Zou

We study the problem of off-policy value evaluation in reinforcement learning (RL), where one aims to estimate the value of a new policy based on data collected by a different policy. This problem is often a critical step when applying RL…

Machine Learning · Computer Science 2016-05-27 Nan Jiang , Lihong Li

In this paper, we establish the global optimality and convergence rate of an off-policy actor critic algorithm in the tabular setting without using density ratio to correct the discrepancy between the state distribution of the behavior…

Machine Learning · Computer Science 2025-02-07 Shangtong Zhang , Remi Tachet , Romain Laroche

A variety of theoretically-sound policy gradient algorithms exist for the on-policy setting due to the policy gradient theorem, which provides a simplified form for the gradient. The off-policy setting, however, has been less clear due to…

Machine Learning · Computer Science 2023-04-17 Eric Graves , Ehsan Imani , Raksha Kumaraswamy , Martha White

The actor-critic (AC) framework has achieved strong empirical success in off-policy reinforcement learning but suffers from the "moving target" problem, where the evaluated policy changes continually. Functional critics, or…

Machine Learning · Computer Science 2026-02-10 Qinxun Bai , Yuxuan Han , Wei Xu , Zhengyuan Zhou

We study a new two-time-scale stochastic gradient method for solving optimization problems, where the gradients are computed with the aid of an auxiliary variable under samples generated by time-varying MDPs controlled by the underlying…

Optimization and Control · Mathematics 2024-08-27 Sihan Zeng , Thinh T. Doan , Justin Romberg

We establish an optimal sample complexity of $O(\epsilon^{-2})$ for obtaining an $\epsilon$-optimal global policy using a single-timescale actor-critic (AC) algorithm in infinite-horizon discounted Markov decision processes (MDPs) with…

Machine Learning · Computer Science 2026-05-08 Navdeep Kumar , Tehila Dahan , Lior Cohen , Ananyabrata Barua , Giorgia Ramponi , Kfir Yehuda Levy , Shie Mannor

We address the discounted reward setting in reinforcement learning (RL). To mitigate the value approximation challenges in policy gradient methods, actor-critic approaches have been developed and are known to converge to stationary points…

Machine Learning · Computer Science 2026-05-15 Sanjeev Manivannan , Shuban V

We propose a fully distributed actor-critic architecture, named Diff-DAC, with application to multitask reinforcement learning (MRL). During the learning process, agents communicate their value and policy parameters to their neighbours,…

Machine Learning · Computer Science 2021-10-26 Sergio Valcarcel Macua , Ian Davies , Aleksi Tukiainen , Enrique Munoz de Cote

We study offline reinforcement learning in average-reward MDPs, which presents increased challenges from the perspectives of distribution shift and non-uniform coverage, and has been relatively underexamined from a theoretical perspective.…

Machine Learning · Computer Science 2026-04-23 Matthew Zurek , Guy Zamir , Yudong Chen

We present the first class of policy-gradient algorithms that work with both state-value and policy function-approximation, and are guaranteed to converge under off-policy training. Our solution targets problems in reinforcement learning…

Artificial Intelligence · Computer Science 2018-02-23 Hamid Reza Maei

Robotic manipulation remains challenging for reinforcement learning due to contact-rich dynamics, long horizons, and training instability. Although off-policy actor-critic algorithms such as SAC and TD3 perform well in simulation, they…

Robotics · Computer Science 2026-04-03 Yu Zhang , Karl Mason

Policy iteration is one of the classical frameworks of reinforcement learning, which requires a known initial stabilizing control. However, finding the initial stabilizing control depends on the known system model. To relax this requirement…

Systems and Control · Electrical Eng. & Systems 2025-03-20 Dongdong Li , Jiuxiang Dong

Reinforcement learning algorithms are typically geared towards optimizing the expected return of an agent. However, in many practical applications, low variance in the return is desired to ensure the reliability of an algorithm. In this…

Machine Learning · Computer Science 2021-02-04 Arushi Jain , Gandharv Patil , Ayush Jain , Khimya Khetarpal , Doina Precup