English
Related papers

Related papers: Temporal Difference Learning as Gradient Splitting

200 papers

In this paper we provide a rigorous convergence analysis of a "off"-policy temporal difference learning algorithm with linear function approximation and per time-step linear computational complexity in "online" learning environment. The…

Machine Learning · Computer Science 2016-05-20 Prasenjit Karmakar , Rajkumar Maity , Shalabh Bhatnagar

Learning rate is a critical component of reinforcement learning (RL). This work uses global and local clocks to distinguish two types of learning rates. The former is of the standard form $\alpha_t$ that depends only on the time step $t$…

Machine Learning · Computer Science 2026-05-11 David Antrobius , Shangtong Zhang

In a broad class of reinforcement learning applications, stochastic rewards have heavy-tailed distributions, which lead to infinite second-order moments for stochastic (semi)gradients in policy evaluation and direct policy optimization. In…

Machine Learning · Computer Science 2023-06-21 Semih Cayci , Atilla Eryilmaz

The concept of the value-gradient is introduced and developed in the context of reinforcement learning. It is shown that by learning the value-gradients exploration or stochastic behaviour is no longer needed to find locally optimal…

Neural and Evolutionary Computing · Computer Science 2008-03-26 Michael Fairbank

Policy evaluation in reinforcement learning is often conducted using two-timescale stochastic approximation, which results in various gradient temporal difference methods such as GTD(0), GTD2, and TDC. Here, we provide convergence rate…

Machine Learning · Computer Science 2019-12-05 Gal Dalal , Balazs Szorenyi , Gugan Thoppe

We consider off-policy temporal-difference (TD) learning in discounted Markov decision processes, where the goal is to evaluate a policy in a model-free way by using observations of a state process generated without executing the policy. To…

Machine Learning · Computer Science 2018-11-27 Huizhen Yu , A. Rupam Mahmood , Richard S. Sutton

Differential temporal difference (TD) methods are value-based reinforcement learning algorithms that have been proposed for infinite-horizon problems. They rely on reward centering, where each reward is centered by the average reward. This…

Machine Learning · Computer Science 2026-05-07 Kris De Asis , Mohamed Elsayed , Jiamin He

We introduce a generalization of temporal-difference (TD) learning to networks of interrelated predictions. Rather than relating a single prediction to itself at a later time, as in conventional TD methods, a TD network relates each…

Machine Learning · Computer Science 2015-04-22 Richard S. Sutton , Brian Tanner

Off-policy learning allows us to learn about possible policies of behavior from experience generated by a different behavior policy. Temporal difference (TD) learning algorithms can become unstable when combined with function approximation…

Machine Learning · Computer Science 2021-06-23 Ray Jiang , Tom Zahavy , Zhongwen Xu , Adam White , Matteo Hessel , Charles Blundell , Hado van Hasselt

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

Policy gradient methods are among the most effective methods in challenging reinforcement learning problems with large state and/or action spaces. However, little is known about even their most basic theoretical convergence properties,…

Machine Learning · Computer Science 2020-10-16 Alekh Agarwal , Sham M. Kakade , Jason D. Lee , Gaurav Mahajan

Predicting and reasoning about the future lie at the heart of many time-series questions. For example, goal-conditioned reinforcement learning can be viewed as learning representations to predict which states are likely to be visited in the…

Machine Learning · Computer Science 2025-10-10 Chongyi Zheng , Ruslan Salakhutdinov , Benjamin Eysenbach

Integral to recent successes in deep reinforcement learning has been a class of temporal difference methods that use infrequently updated target values for policy evaluation in a Markov Decision Process. Yet a complete theoretical…

Machine Learning · Computer Science 2023-08-15 Mattie Fellows , Matthew J. A. Smith , Shimon Whiteson

Temporal difference (TD) learning is a cornerstone of reinforcement learning. In the average-reward setting, standard TD($\lambda$) is highly sensitive to the choice of step-size and thus requires careful tuning to maintain numerical…

Machine Learning · Statistics 2025-10-08 Hwanwoo Kim , Dongkyu Derek Cho , Eric Laber

Balancing between computational efficiency and sample efficiency is an important goal in reinforcement learning. Temporal difference (TD) learning algorithms stochastically update the value function, with a linear time complexity in the…

Machine Learning · Computer Science 2016-11-21 Clement Gehring , Yangchen Pan , Martha White

We study the problem of computing the value function from a discretely-observed trajectory of a continuous-time diffusion process. We develop a new class of algorithms based on easily implementable numerical schemes that are compatible with…

Machine Learning · Computer Science 2024-07-09 Wenlong Mou , Yuhua Zhu

We derive a family of risk-sensitive reinforcement learning methods for agents, who face sequential decision-making tasks in uncertain environments. By applying a utility function to the temporal difference (TD) error, nonlinear…

Machine Learning · Computer Science 2014-10-10 Yun Shen , Michael J. Tobia , Tobias Sommer , Klaus Obermayer

Processes (MDPs) often require frequent decision making, that is, taking an action every microsecond, second, or minute. Infinite horizon discount reward formulation is still relevant for a large portion of these applications, because…

Optimization and Control · Mathematics 2014-12-17 Yin-Lam Chow , Junjie Qin

The policy gradient theorem (Sutton et al., 2000) prescribes the usage of a cumulative discounted state distribution under the target policy to approximate the gradient. Most algorithms based on this theorem, in practice, break this…

Machine Learning · Computer Science 2022-07-08 Samuele Tosatto , Andrew Patterson , Martha White , A. Rupam Mahmood

Gradient-based temporal difference (GTD) algorithms are widely used in off-policy learning scenarios. Among them, the two time-scale TD with gradient correction (TDC) algorithm has been shown to have superior performance. In contrast to…

Machine Learning · Computer Science 2019-09-27 Tengyu Xu , Shaofeng Zou , Yingbin Liang