Reinforcement Learning: An Introduction, Second Edition (Richard S. Sutton, et al)

 
0.0 (0)
Reinforcement Learning: An Introduction, Second Edition (Richard S. Sutton, et al)

In Reinforcement Learning (RL), one of the most active study topics in artificial intelligence, an agent interacts with a complex, uncertain environment while attempting to maximize the overall number of rewards it receives.

The main concepts and algorithms of the area are outlined in Reinforcement Learning by Richard Sutton and Andrew Barto. This second version has been greatly enlarged and revised, adding new themes and revising the treatment of existing ones.

The more mathematical content is highlighted in shaded boxes in this second edition, which, like the first, concentrates on fundamental online learning methods. All readers in related subjects should be able to understand the treatment.

Without moving beyond the tabular situation for which exact answers may be obtained, Part I covers as much reinforcement learning as it can. In the second edition, several of the algorithms discussed in this section are brand-new, such as UCB, Expected Sarsa, and Double Learning.

Part II expands these concepts to function approximation, provides extensive coverage of off-policy learning and policy-gradient techniques, and adds new sections on topics like artificial neural networks and the Fourier basis.

Part III includes brand-new chapters on the connections between psychology and neuroscience and reinforcement learning, as well as an updated case-study chapter that includes playing AlphaGo and AlphaGo Zero, playing Atari games, and IBM Watson's betting strategy. The last chapter explores how reinforcement learning will affect society in the future.

This book is considered to be the holy grail of reinforcement learning, and the release of the new edition coincides perfectly with the field's exploding activity. No student, researcher, practitioner, or inquisitive nonspecialist who is interested in the issue of learning to act should be without it.

Ebook Details

About the Authors
  • Richard S. Sutton is a Canadian computer scientist who holds the position of Distinguished Research Scientist at DeepMind as well as Professor of Computing Science and AITF Chair in Reinforcement Learning and Artificial Intelligence at the University of Alberta.
  • Professor of computer science at the University of Massachusetts Amherst and department chair since January 2007, Andrew G. Barto Reinforcement learning is his primary field of study.
Published
Published Date / Year
2018; eBook (Creative Commons Licensed)
License(s)
CC BY-NC-ND 2.0
Hardcover
522 pages
eBook Format
PDF (548 pages)
ISBN-10
0262039249
ISBN-13
978-0262039246

Similar Programming & Computer Books

Greasemonkey Hacks: Tips & Tools for Remixing the Web with Firefox (Mark Pilgrim)
For hardcore users who wish to learn Greasemonkey, the hottest new Firefox plugin that enables you to write scripts that modify the web pages you see, this book is...
Strategic Foundations of General Equilibrium: Dynamic Matching and Bargaining Games (Douglas Gale)
Since Adam Smith's day, the theory of competition has played a significant role in economic study. This book, published by one of the most eminent modern economic theorists, details...
The Pure Logic Of Choice (Richard D. Fuerle)
A broad theory of economics based on free will is presented in this free programming book. The assumption that humans have free will and the ability to alter physical...
Portfolio Theory and Financial Analyses (Robert Alan Hill)
Whether they involve calculating the return on a portfolio, analyzing portfolio risk, or assessing the effectiveness of the portfolio management process, this free programming book links each of the...
Price Theory: An Intermediate Text (David D. Friedman)
In order to help the reader grasp the economic way of thinking, the author first gives verbal, intuitive explanations of the topics before using graphs and/or calculus to illustrate...
Mathematical Models in Portfolio Analysis (Farida Kachapova)
This free programming book presents the mathematical theory of portfolio modeling in financial mathematics as a coherent whole, with justifications for each step. ...
Stochastic Calculus and Finance (Steven E. Shreve)
The first 10 years of the Carnegie Mellon Professional Master program in Computational Finance led to the development of stochastic calculus for finance. Students with calculus and probability based...
Math for Trades: Volume 1 (Chad Flinn, et al.)
The foundational elements for learning math are presented in this volume. Whole numbers, fractions, decimals, and percents are all included in the book. ...
Handbook of Digital Face Manipulation and Detection: From DeepFakes to Morphing Attacks (Christian Rathgeb, et al)
The first thorough compilation of research on the popular subject of digital face alteration, including DeepFakes, Face Morphing, and Reenactment, is offered in this open access book. ...
Multimedia Fingerprinting Forensics for Traitor Tracing (K. J. Ray Liu, et al)
The widespread dissemination and consumption of digital multimedia data is a result of the appeal of multimedia content.

User reviews

There are no user reviews for this listing.
Ratings
Rate this Book
Comments