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

Sutton, Richard S.; Barto, Andrew G.

ISBN 10: 0262193981 / ISBN 13: 9780262193986

Published by Bradford Books, 1998

About this Item

Condition: 15 pages with some highlighted text, the rest excellent.

The book provides a clear and simple account of the key ideas and algorithms in this area of artificial intelligence, where an agent learns to maximize a cumulative reward by interacting with a complex, uncertain environment.

It covers the history of the field's intellectual foundations and proceeds to the core algorithms and concepts, including:

The Reinforcement Learning Problem framed in terms of Markov Decision Processes (MDPs).

Basic Solution Methods like Dynamic Programming, Monte Carlo methods, and the influential Temporal-Difference (TD) learning (e.g., Q-learning and SARSA).

Function Approximation for handling large state spaces, including the use of artificial neural networks.

More advanced topics like policy-gradient methods and a discussion of RL's relationships to psychology and neuroscience.

Often referred to as the "bible" of the field, it is a foundational text suitable for students, researchers, and practitioners with a basic understanding of probability.

Seller Inventory # ABE-1760107744142

Bibliographic Details

• Title: Reinforcement Learning: An Introduction (Adaptive Computation and Machine Learning)
• Publisher: Bradford Books
• Publication Date: 1998
• Language: English
• ISBN 10: 0262193981
• ISBN 13: 9780262193986
• Binding: Hardcover
• Edition: 1st Edition
• Printing: 1st Printing
• Condition: Very Good
• Dimensions: 3.175W x 18.415H x 24.13D centimeters 3.175 centimeters width by 18.415 centimeters height by 24.13 centimeters depth
• Weight: 794 grams
• Seller catalogs:
  • Reinforcement learning
  • Data science
  • Machine Learning

About this title

Synopsis

Richard Sutton and Andrew Barto provide a clear and simple account of the key ideas and algorithms of reinforcement learning. Their discussion ranges from the history of the field's intellectual foundations to the most recent developments and applications.

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 when interacting with a complex, uncertain environment. In Reinforcement Learning, Richard Sutton and Andrew Barto provide a clear and simple account of the key ideas and algorithms of reinforcement learning. Their discussion ranges from the history of the field's intellectual foundations to the most recent developments and applications. The only necessary mathematical background is familiarity with elementary concepts of probability.

The book is divided into three parts. Part I defines the reinforcement learning problem in terms of Markov decision processes. Part II provides basic solution methods: dynamic programming, Monte Carlo methods, and temporal-difference learning. Part III presents a unified view of the solution methods and incorporates artificial neural networks, eligibility traces, and planning; the two final chapters present case studies and consider the future of reinforcement learning.

About the Authors

Andrew G. Barto is Professor of Computer Science at the University of Massachusetts.

Richard S. Sutton is Senior Research Scientist, Department of Computer Science, University of Massachusetts.