The significantly expanded and updated new edition of a widely used text on reinforcement learning, one of the most active research areas in artificial intelligence. Reinforcement learning, one of the most active research areas in artificial intelligence, is a computational approach to learning whereby an agent tries to maximize the total amount of reward it receives while interacting with a complex, uncertain environment. In Reinforcement Learning , Richard Sutton and Andrew Barto provide a clear and simple account of the field's key ideas and algorithms. This second edition has been significantly expanded and updated, presenting new topics and updating coverage of other topics.Like the first edition, this second edition focuses on core online learning algorithms, with the more mathematical material set off in shaded boxes. Part I covers as much of reinforcement learning as possible without going beyond the tabular case for which exact solutions can be found. Many algorithms presented in this part are new to the second edition, including UCB, Expected Sarsa, and Double Learning. Part II extends these ideas to function approximation, with new sections on such topics as artificial neural networks and the Fourier basis, and offers expanded treatment of off-policy learning and policy-gradient methods. Part III has new chapters on reinforcement learning's relationships to psychology and neuroscience, as well as an updated case-studies chapter including AlphaGo and AlphaGo Zero, Atari game playing, and IBM Watson's wagering strategy. The final chapter discusses the future societal impacts of reinforcement learning.

lgrsnf/强化学习导论中文第二版.pdf

Reinforcement Learning: An Introduction (Adaptive Computation and Machine Learning series)

Reinforcement Learning, Second Edition : An Introduction

Обучение с подкреплением: введение

Ричард С. Саттон, Эндрю Дж. Барто; перевод с английского А. А. Слинкина

Sutton, Richard S., Barto, Andrew G.

Richard S. Sutton; Andrew G. Barto

Саттон, Ричард С

A Bradford Book

Bradford Books

The MIT Press

AAAI Press

ДМК Пресс

Adaptive computation and machine learning, Second edition, Cambridge, Massachusetts, 2018

Adaptive computation and machine learning, 2018: 1, second edition, Cambridge, Mass, 2018

MIT Press, Cambridge, Massachusetts, 2018

United States, United States of America

2-е изд., Москва, Russia, 2020

Nov 13, 2018

Source title: Reinforcement Learning: An Introduction (Adaptive Computation and Machine Learning series)

На обл.: Цифра
Предм. указ.: с. 546-551
Пер.: Sutton, Richard S. Reinforcement learning 2nd ed. 978-0-262-03924-6

РГБ

Russian State Library [rgb] MARC:
=001 010417813
=005 20210423100821.0
=008 200713s2020\\\\ru\\\\\\\\\\\\000\0\rus\d
=017 \\ $a 7125-20 $b RuMoRGB
=020 \\ $a 978-5-97060-097-9 $c 500 экз.
=040 \\ $a RuMoRGB $b rus $e rcr
=041 1\ $a rus $h eng
=044 \\ $a ru
=084 \\ $a З818.6,0 $2 rubbk
=084 \\ $a З813.5,0 $2 rubbk
=084 \\ $a Ч30в2:З81,0 $2 rubbk
=100 1\ $a Саттон, Ричард С.
=245 00 $a Обучение с подкреплением : $b введение $c Ричард С. Саттон, Эндрю Дж. Барто ; перевод с английского А. А. Слинкина
=250 \\ $a 2-е изд.
=260 \\ $a Москва $b ДМК Пресс $c 2020
=300 \\ $a 551 с. $b ил., цв. ил. $c 24 см
=336 \\ $a Текст (визуальный)
=337 \\ $a непосредственный
=500 \\ $a На обл.: Цифра
=500 \\ $a Предм. указ.: с. 546-551
=534 \\ $p Пер.: $a Sutton, Richard S. $t Reinforcement learning $b 2nd ed. $z 978-0-262-03924-6
=650 \7 $a Радиоэлектроника -- Кибернетика -- Бионика -- Биоэлектрические модели -- Нейронные сети $2 rubbk
=650 \7 $a Радиоэлектроника -- Кибернетика -- Искусственный интеллект -- Системы искусственного интеллекта $2 rubbk
=650 \7 $a Культура. Наука. Просвещение -- Народное образование. Педагогические науки -- Педагогика -- Методология педагогики -- Педагогика и другие науки -- Педагогика и кибернетика $2 rubbk
=700 1\ $a Барто, Эндрю Дж. $d 1948-
=852 4\ $a РГБ $b FB $j 2 20-47/121 $x 90

Reinforcement learning is the learning of a mapping from situations to actions so as to maximize a scalar reward or reinforcement signal. The learner is not told which action to take, as in most forms of machine learning, but instead must discover which actions yield the highest reward by trying them. In the most interesting and challenging cases, actions may affect not only the immediate reward, but also the next situation, and through that all subsequent rewards. These two characteristics -- trial-and-error search and delayed reward -- are the most important distinguishing features of reinforcement learning. Reinforcement learning is both a new and a very old topic in AI. The term appears to have been coined by Minsk (1961), and independently in control theory by Walz and Fu (1965). The earliest machine learning research now viewed as directly relevant was Samuel's (1959) checker player, which used temporal-difference learning to manage delayed reward much as it is used today. Of course learning and reinforcement have been studied in psychology for almost a century, and that work has had a very strong impact on the AI/engineering work. One could in fact consider all of reinforcement learning to be simply the reverse engineering of certain psychological learning processes (e.g. operant conditioning and secondary reinforcement). Reinforcement Learning is an edited volume of original research, comprising seven invited contributions by leading researchers.

2024-02-07