English
Related papers

Related papers: O$^2$TD: (Near)-Optimal Off-Policy TD Learning

200 papers

The use of target networks has been a popular and key component of recent deep Q-learning algorithms for reinforcement learning, yet little is known from the theory side. In this work, we introduce a new family of target-based temporal…

Machine Learning · Computer Science 2019-09-24 Donghwan Lee , Niao He

Learning the value function of a given policy from data samples is an important problem in Reinforcement Learning. TD($\lambda$) is a popular class of algorithms to solve this problem. However, the weights assigned to different $n$-step…

Machine Learning · Computer Science 2021-11-24 Rohan Deb , Meet Gandhi , Shalabh Bhatnagar

Using insight from numerical approximation of ODEs and the problem formulation and solution methodology of TD learning through a Galerkin relaxation, I propose a new class of TD learning algorithms. After applying the improved numerical…

Machine Learning · Computer Science 2021-04-21 Caleb Bowyer

The family of temporal difference (TD) methods span a spectrum from computationally frugal linear methods like TD({\lambda}) to data efficient least squares methods. Least square methods make the best use of available data directly…

Artificial Intelligence · Computer Science 2017-03-13 Yangchen Pan , Adam White , Martha White

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

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

The goal of this manuscript is to conduct a controltheoretic analysis of Temporal Difference (TD) learning algorithms. TD-learning serves as a cornerstone in the realm of reinforcement learning, offering a methodology for approximating the…

Artificial Intelligence · Computer Science 2023-09-12 Donghwan Lee , Do Wan Kim

Gradient descent or its variants are popular in training neural networks. However, in deep Q-learning with neural network approximation, a type of reinforcement learning, gradient descent (also known as Residual Gradient (RG)) is barely…

Machine Learning · Computer Science 2022-11-15 Shuyu Yin , Tao Luo , Peilin Liu , Zhi-Qin John Xu

A natural optimization model that formulates many online resource allocation and revenue management problems is the online linear program (LP) in which the constraint matrix is revealed column by column along with the corresponding…

Data Structures and Algorithms · Computer Science 2014-04-10 Shipra Agrawal , Zizhuo Wang , Yinyu Ye

Temporal difference (TD) learning is one of the main foundations of modern reinforcement learning. This paper studies the use of TD(0), a canonical TD algorithm, to estimate the value function of a given policy from a batch of data. In this…

Machine Learning · Computer Science 2020-08-18 Brahma Pavse , Ishan Durugkar , Josiah Hanna , Peter Stone

Policy gradient methods are an appealing approach in reinforcement learning because they directly optimize the cumulative reward and can straightforwardly be used with nonlinear function approximators such as neural networks. The two main…

Machine Learning · Computer Science 2018-10-23 John Schulman , Philipp Moritz , Sergey Levine , Michael Jordan , Pieter Abbeel

We study the convergence of off-policy TD(0) with linear function approximation when used to approximate the expected discounted reward in a Markov chain. It is well known that the combination of off-policy learning and function…

Machine Learning · Statistics 2026-04-02 Maik Overmars , Jasper Goseling , Richard Boucherie

Reinforcement Learning aims at identifying and evaluating efficient control policies from data. In many real-world applications, the learner is not allowed to experiment and cannot gather data in an online manner (this is the case when…

Machine Learning · Computer Science 2024-07-02 Daniele Foffano , Alessio Russo , Alexandre Proutiere

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

Recent advances of gradient temporal-difference methods allow to learn off-policy multiple value functions in parallel with- out sacrificing convergence guarantees or computational efficiency. This opens up new possibilities for sound…

Artificial Intelligence · Computer Science 2014-05-22 Anna Harutyunyan , Tim Brys , Peter Vrancx , Ann Nowe

TD($\lambda$) with function approximation has proved empirically successful for some complex reinforcement learning problems. For linear approximation, TD($\lambda$) has been shown to minimise the squared error between the approximate value…

Machine Learning · Computer Science 2025-12-24 Lex Weaver , Jonathan Baxter

Fast-converging algorithms are a contemporary requirement in reinforcement learning. In the context of linear function approximation, the magnitude of the smallest eigenvalue of the key matrix is a major factor reflecting the convergence…

Machine Learning · Computer Science 2024-11-12 Xingguo Chen , Yu Gong , Shangdong Yang , Wenhao Wang

When performing imitation learning from expert demonstrations, distribution matching is a popular approach, in which one alternates between estimating distribution ratios and then using these ratios as rewards in a standard reinforcement…

Machine Learning · Computer Science 2019-12-12 Ilya Kostrikov , Ofir Nachum , Jonathan Tompson

Reinforcement learning algorithms typically rely on the assumption that the environment dynamics and value function can be expressed in terms of a Markovian state representation. However, when state information is only partially observable,…

Learning rate schedules used in practice bear little resemblance to those recommended by theory. We close much of this theory/practice gap, and as a consequence are able to derive new problem-adaptive learning rate schedules. Our main…

Machine Learning · Computer Science 2024-10-31 Aaron Defazio , Ashok Cutkosky , Harsh Mehta , Konstantin Mishchenko