Overview
- Latest research on Temporal Difference Reinforcement Learning for Robots
- Focuses on applying Reinforcement Learning to real-world problems, particularly learning on robots
- Presents the model-based Reinforcement Learning algorithm developed by the authors group
- Written by an expert in the field
Part of the book series: Studies in Computational Intelligence (SCI, volume 503)
Buy print copy
About this book
This book presents and develops new reinforcement learning methods that enable fast and robust learning on robots in real-time.
Robots have the potential to solve many problems in society, because of their ability to work in dangerous places doing necessary jobs that no one wants or is able to do. One barrier to their widespread deployment is that they are mainly limited to tasks where it is possible to hand-program behaviors for every situation that may be encountered. For robots to meet their potential, they need methods that enable them to learn and adapt to novel situations that they were not programmed for. Reinforcement learning (RL) is a paradigm for learning sequential decision making processes and could solve the problems of learning and adaptation on robots. This book identifies four key challenges that must be addressed for an RL algorithm to be practical for robotic control tasks. These RL for Robotics Challenges are: 1) it must learn in very few samples; 2) it must learn in domains with continuous state features; 3) it must handle sensor and/or actuator delays; and 4) it should continually select actions in real time. This book focuses on addressing all four of these challenges. In particular, this book is focused on time-constrained domains where the first challenge is critically important. In these domains, the agent’s lifetime is not long enough for it to explore the domains thoroughly, and it must learn in very few samples.
Similar content being viewed by others
Keywords
Table of contents (8 chapters)
Authors and Affiliations
Bibliographic Information
Book Title: TEXPLORE: Temporal Difference Reinforcement Learning for Robots and Time-Constrained Domains
Authors: Todd Hester
Series Title: Studies in Computational Intelligence
DOI: https://doi.org/10.1007/978-3-319-01168-4
Publisher: Springer Cham
eBook Packages: Engineering, Engineering (R0)
Copyright Information: Springer International Publishing Switzerland 2013
Hardcover ISBN: 978-3-319-01167-7Published: 04 July 2013
Softcover ISBN: 978-3-319-37510-6Published: 24 September 2016
eBook ISBN: 978-3-319-01168-4Published: 22 June 2013
Series ISSN: 1860-949X
Series E-ISSN: 1860-9503
Edition Number: 1
Number of Pages: XIV, 165
Number of Illustrations: 55 illustrations in colour
Topics: Computational Intelligence, Image Processing and Computer Vision, Robotics and Automation