English
Related papers

Related papers: Why Target Networks Stabilise Temporal Difference …

200 papers

Off-policy reinforcement learning has many applications including: learning from demonstration, learning multiple goal seeking policies in parallel, and representing predictive knowledge. Recently there has been an proliferation of new…

Machine Learning · Computer Science 2016-04-01 Adam White , Martha White

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

We consider the core reinforcement-learning problem of on-policy value function approximation from a batch of trajectory data, and focus on various issues of Temporal Difference (TD) learning and Monte Carlo (MC) policy evaluation. The two…

Q-learning played a foundational role in the field reinforcement learning (RL). However, TD algorithms with off-policy data, such as Q-learning, or nonlinear function approximation like deep neural networks require several additional tricks…

Machine Learning · Computer Science 2025-04-23 Matteo Gallici , Mattie Fellows , Benjamin Ellis , Bartomeu Pou , Ivan Masmitja , Jakob Nicolaus Foerster , Mario Martin

We revisit residual algorithms in both model-free and model-based reinforcement learning settings. We propose the bidirectional target network technique to stabilize residual algorithms, yielding a residual version of DDPG that…

Machine Learning · Computer Science 2020-01-27 Shangtong Zhang , Wendelin Boehmer , Shimon Whiteson

Temporal-Difference (TD) learning is a standard and very successful reinforcement learning approach, at the core of both algorithms that learn the value of a given policy, as well as algorithms which learn how to improve policies.…

Machine Learning · Computer Science 2020-05-19 Mingde Zhao , Sitao Luan , Ian Porada , Xiao-Wen Chang , Doina Precup

We know from reinforcement learning theory that temporal difference learning can fail in certain cases. Sutton and Barto (2018) identify a deadly triad of function approximation, bootstrapping, and off-policy learning. When these three…

Artificial Intelligence · Computer Science 2018-12-07 Hado van Hasselt , Yotam Doron , Florian Strub , Matteo Hessel , Nicolas Sonnerat , Joseph Modayil

Neural Temporal Difference (TD) Learning is an approximate temporal difference method for policy evaluation that uses a neural network for function approximation. Analysis of Neural TD Learning has proven to be challenging. In this paper we…

Machine Learning · Computer Science 2023-12-12 Haoxing Tian , Ioannis Ch. Paschalidis , Alex Olshevsky

Adversarial neural networks solve many important problems in data science, but are notoriously difficult to train. These difficulties come from the fact that optimal weights for adversarial nets correspond to saddle points, and not…

Machine Learning · Computer Science 2018-02-12 Abhay Yadav , Sohil Shah , Zheng Xu , David Jacobs , Tom Goldstein

One of the most basic problems in reinforcement learning (RL) is policy evaluation: estimating the long-term return, i.e., value function, corresponding to a given fixed policy. The celebrated Temporal Difference (TD) learning algorithm…

Machine Learning · Computer Science 2025-02-10 Sreejeet Maity , Aritra Mitra

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

The primary goal of reinforcement learning is to develop decision-making policies that prioritize optimal performance, frequently without considering safety. In contrast, safe reinforcement learning seeks to reduce or avoid unsafe behavior.…

Machine Learning · Computer Science 2025-06-17 Zahra Shahrooei , Ali Baheri

This paper analyzes multi-step temporal difference (TD)-learning algorithms within the ``deadly triad'' scenario, characterized by linear function approximation, off-policy learning, and bootstrapping. In particular, we prove that $n$-step…

Machine Learning · Computer Science 2026-02-24 Han-Dong Lim , Donghwan Lee

We devise a distributional variant of gradient temporal-difference (TD) learning. Distributional reinforcement learning has been demonstrated to outperform the regular one in the recent study \citep{bellemare2017distributional}. In the…

Machine Learning · Computer Science 2019-04-04 Chao Qu , Shie Mannor , Huan Xu

Larger and deeper networks generalise well despite their increased capacity to overfit. Understanding why this happens is theoretically and practically important. One recent approach looks at the infinitely wide limits of such networks and…

Machine Learning · Computer Science 2023-10-13 Adrian Goldwaser , Hong Ge

Reinforcement learning in non-stationary environments is challenging due to abrupt and unpredictable changes in dynamics, often causing traditional algorithms to fail to converge. However, in many real-world cases, non-stationarity has some…

Machine Learning · Computer Science 2025-03-25 Mohsen Amiri , Sindri Magnússon

Actor-critic methods integrating target networks have exhibited a stupendous empirical success in deep reinforcement learning. However, a theoretical understanding of the use of target networks in actor-critic methods is largely missing in…

Machine Learning · Computer Science 2022-02-24 Anas Barakat , Pascal Bianchi , Julien Lehmann

Multi-step temporal-difference (TD) learning, where the update targets contain information from multiple time steps ahead, is one of the most popular forms of TD learning for linear function approximation. The reason is that multi-step…

Artificial Intelligence · Computer Science 2016-08-19 Harm van Seijen

Most reinforcement learning algorithms implicitly assume strong synchrony. We present novel attacks targeting Q-learning that exploit a vulnerability entailed by this assumption by delaying the reward signal for a limited time period. We…

Machine Learning · Computer Science 2022-09-09 Anindya Sarkar , Jiarui Feng , Yevgeniy Vorobeychik , Christopher Gill , Ning Zhang

Temporal-Difference (TD) learning methods, such as Q-Learning, have proven effective at learning a policy to perform control tasks. One issue with methods like Q-Learning is that the value update introduces bias when predicting the TD…

Machine Learning · Computer Science 2021-10-29 Litian Liang , Yaosheng Xu , Stephen McAleer , Dailin Hu , Alexander Ihler , Pieter Abbeel , Roy Fox