English
Related papers

Related papers: Convergence of off-policy TD(0) with linear functi…

200 papers

Model-free algorithms for reinforcement learning typically require a condition called Bellman completeness in order to successfully operate off-policy with function approximation, unless additional conditions are met. However, Bellman…

Machine Learning · Computer Science 2023-06-07 Andrea Zanette

Temporal-difference learning with gradient correction (TDC) is a two time-scale algorithm for policy evaluation in reinforcement learning. This algorithm was initially proposed with linear function approximation, and was later extended to…

Machine Learning · Computer Science 2021-10-29 Yue Wang , Shaofeng Zou , Yi Zhou

Off-policy algorithms, in which a behavior policy differs from the target policy and is used to gain experience for learning, have proven to be of great practical value in reinforcement learning. However, even for simple convex problems…

Machine Learning · Computer Science 2022-09-13 Rong J. B. Zhu , James M. Murray

We present for the first time an asymptotic convergence analysis of two time-scale stochastic approximation driven by "controlled" Markov noise. In particular, the faster and slower recursions have non-additive controlled Markov noise…

Machine Learning · Computer Science 2020-12-03 Prasenjit Karmakar

In this paper, we study the finite-sample statistical rates of distributional temporal difference (TD) learning with linear function approximation. The aim of distributional TD learning is to estimate the return distribution of a discounted…

Machine Learning · Statistics 2025-05-14 Yang Peng , Kaicheng Jin , Liangyu Zhang , Zhihua Zhang

We establish a connection between policy evaluation in Markov decision processes and PageRank in network analysis. For a fixed policy, we show that the value function of a discounted Markov decision process can be obtained, up to an…

Optimization and Control · Mathematics 2026-05-04 Konstantin Avrachenkov , Lorenzo Gregoris , Nelly Litvak

We consider off-policy evaluation of dynamic treatment rules under sequential ignorability, given an assumption that the underlying system can be modeled as a partially observed Markov decision process (POMDP). We propose an estimator,…

Machine Learning · Computer Science 2023-05-10 Yuchen Hu , Stefan Wager

We establish novel and general high-dimensional concentration inequalities and Berry-Esseen bounds for vector-valued martingales induced by Markov chains. We apply these results to analyze the performance of the Temporal Difference (TD)…

Machine Learning · Statistics 2026-05-22 Weichen Wu , Yuting Wei , Alessandro Rinaldo

Off-policy learning methods seek to derive an optimal policy directly from a fixed dataset of prior interactions. This objective presents significant challenges, primarily due to the inherent distributional shift and value function…

Machine Learning · Computer Science 2026-02-03 Arip Asadulaev , Maksim Bobrin , Salem Lahlou , Dmitry Dylov , Fakhri Karray , Martin Takac

In deep reinforcement learning, policy optimization methods need to deal with issues such as function approximation and the reuse of off-policy data. Standard policy gradient methods do not handle off-policy data well, leading to premature…

Machine Learning · Computer Science 2025-01-28 Qing Wang , Yingru Li , Jiechao Xiong , Tong Zhang

We consider the emphatic temporal-difference (TD) algorithm, ETD($\lambda$), for learning the value functions of stationary policies in a discounted, finite state and action Markov decision process. The ETD($\lambda$) algorithm was recently…

Machine Learning · Computer Science 2017-01-23 Huizhen Yu

Temporal difference learning (TD) is a simple iterative algorithm used to estimate the value function corresponding to a given policy in a Markov decision process. Although TD is one of the most widely used algorithms in reinforcement…

Machine Learning · Computer Science 2018-11-07 Jalaj Bhandari , Daniel Russo , Raghav Singal

Policy gradient methods in reinforcement learning update policy parameters by taking steps in the direction of an estimated gradient of policy value. In this paper, we consider the statistically efficient estimation of policy gradients from…

Machine Learning · Statistics 2020-02-21 Nathan Kallus , Masatoshi Uehara

Reinforcement learning with function approximation can be unstable and even divergent, especially when combined with off-policy learning and Bellman updates. In deep reinforcement learning, these issues have been dealt with empirically by…

Machine Learning · Computer Science 2020-10-06 Dibya Ghosh , Marc G. Bellemare

In this paper we consider the problem of obtaining sharp bounds for the performance of temporal difference (TD) methods with linear function approximation for policy evaluation in discounted Markov decision processes. We show that a simple…

Machine Learning · Statistics 2024-06-18 Sergey Samsonov , Daniil Tiapkin , Alexey Naumov , Eric Moulines

Temporal difference learning with linear function approximation is a popular method to obtain a low-dimensional approximation of the value function of a policy in a Markov Decision Process. We give a new interpretation of this method in…

Machine Learning · Computer Science 2020-10-29 Rui Liu , Alex Olshevsky

Large scale reinforcement learning has become a central tool for improving reasoning in large language models. At this scale, generation is often lagged or asynchronous, so updates are performed on data collected by older policies. This…

Machine Learning · Computer Science 2026-05-28 Otmane Sakhi , Aleksei Arzhantsev , Imad Aouali , Flavian Vasile

Temporal-Difference learning (TD) [Sutton, 1988] with function approximation can converge to solutions that are worse than those obtained by Monte-Carlo regression, even in the simple case of on-policy evaluation. To increase our…

Machine Learning · Computer Science 2018-07-10 Hugo Penedones , Damien Vincent , Hartmut Maennel , Sylvain Gelly , Timothy Mann , Andre Barreto

TD(0) is one of the most commonly used algorithms in reinforcement learning. Despite this, there is no existing finite sample analysis for TD(0) with function approximation, even for the linear case. Our work is the first to provide such…

Artificial Intelligence · Computer Science 2017-12-12 Gal Dalal , Balázs Szörényi , Gugan Thoppe , Shie Mannor

We prove that the combination of a target network and over-parameterized linear function approximation establishes a weaker convergence condition for bootstrapped value estimation in certain cases, even with off-policy data. Our condition…