Adaptive Representations for Reinforcement Learning (Record no. 112367)
[ view plain ]
| 000 -LEADER | |
|---|---|
| fixed length control field | 03533nam a22004455i 4500 |
| 001 - CONTROL NUMBER | |
| control field | 978-3-642-13932-1 |
| 003 - CONTROL NUMBER IDENTIFIER | |
| control field | DE-He213 |
| 005 - DATE AND TIME OF LATEST TRANSACTION | |
| control field | 20140220084540.0 |
| 007 - PHYSICAL DESCRIPTION FIXED FIELD--GENERAL INFORMATION | |
| fixed length control field | cr nn 008mamaa |
| 008 - FIXED-LENGTH DATA ELEMENTS--GENERAL INFORMATION | |
| fixed length control field | 100709s2010 gw | s |||| 0|eng d |
| 020 ## - INTERNATIONAL STANDARD BOOK NUMBER | |
| International Standard Book Number | 9783642139321 |
| -- | 978-3-642-13932-1 |
| 024 7# - OTHER STANDARD IDENTIFIER | |
| Standard number or code | 10.1007/978-3-642-13932-1 |
| Source of number or code | doi |
| 050 #4 - LIBRARY OF CONGRESS CALL NUMBER | |
| Classification number | Q342 |
| 072 #7 - SUBJECT CATEGORY CODE | |
| Subject category code | UYQ |
| Source | bicssc |
| 072 #7 - SUBJECT CATEGORY CODE | |
| Subject category code | COM004000 |
| Source | bisacsh |
| 082 04 - DEWEY DECIMAL CLASSIFICATION NUMBER | |
| Classification number | 006.3 |
| Edition number | 23 |
| 100 1# - MAIN ENTRY--PERSONAL NAME | |
| Personal name | Whiteson, Shimon. |
| Relator term | author. |
| 245 10 - TITLE STATEMENT | |
| Title | Adaptive Representations for Reinforcement Learning |
| Medium | [electronic resource] / |
| Statement of responsibility, etc | by Shimon Whiteson. |
| 264 #1 - | |
| -- | Berlin, Heidelberg : |
| -- | Springer Berlin Heidelberg, |
| -- | 2010. |
| 300 ## - PHYSICAL DESCRIPTION | |
| Extent | 133p. 11 illus. in color. |
| Other physical details | online resource. |
| 336 ## - | |
| -- | text |
| -- | txt |
| -- | rdacontent |
| 337 ## - | |
| -- | computer |
| -- | c |
| -- | rdamedia |
| 338 ## - | |
| -- | online resource |
| -- | cr |
| -- | rdacarrier |
| 347 ## - | |
| -- | text file |
| -- | |
| -- | rda |
| 490 1# - SERIES STATEMENT | |
| Series statement | Studies in Computational Intelligence, |
| International Standard Serial Number | 1860-949X ; |
| Volume number/sequential designation | 291 |
| 505 0# - FORMATTED CONTENTS NOTE | |
| Formatted contents note | Part 1 Introduction -- Part 2 Reinforcement Learning -- Part 3 On-Line Evolutionary Computation -- Part 4 Evolutionary Function Approximation -- Part 5 Sample-Efficient Evolutionary Function Approximation -- Part 6 Automatic Feature Selection for Reinforcement Learning -- Part 7 Adaptive Tile Coding -- Part 8 RelatedWork -- Part 9 Conclusion -- Part 10 Statistical Significance. |
| 520 ## - SUMMARY, ETC. | |
| Summary, etc | This book presents new algorithms for reinforcement learning, a form of machine learning in which an autonomous agent seeks a control policy for a sequential decision task. Since current methods typically rely on manually designed solution representations, agents that automatically adapt their own representations have the potential to dramatically improve performance. This book introduces two novel approaches for automatically discovering high-performing representations. The first approach synthesizes temporal difference methods, the traditional approach to reinforcement learning, with evolutionary methods, which can learn representations for a broad class of optimization problems. This synthesis is accomplished by customizing evolutionary methods to the on-line nature of reinforcement learning and using them to evolve representations for value function approximators. The second approach automatically learns representations based on piecewise-constant approximations of value functions. It begins with coarse representations and gradually refines them during learning, analyzing the current policy and value function to deduce the best refinements. This book also introduces a novel method for devising input representations. This method addresses the feature selection problem by extending an algorithm that evolves the topology and weights of neural networks such that it evolves their inputs too. In addition to introducing these new methods, this book presents extensive empirical results in multiple domains demonstrating that these techniques can substantially improve performance over methods with manual representations. |
| 650 #0 - SUBJECT ADDED ENTRY--TOPICAL TERM | |
| Topical term or geographic name as entry element | Engineering. |
| 650 #0 - SUBJECT ADDED ENTRY--TOPICAL TERM | |
| Topical term or geographic name as entry element | Artificial intelligence. |
| 650 14 - SUBJECT ADDED ENTRY--TOPICAL TERM | |
| Topical term or geographic name as entry element | Engineering. |
| 650 24 - SUBJECT ADDED ENTRY--TOPICAL TERM | |
| Topical term or geographic name as entry element | Computational Intelligence. |
| 650 24 - SUBJECT ADDED ENTRY--TOPICAL TERM | |
| Topical term or geographic name as entry element | Artificial Intelligence (incl. Robotics). |
| 710 2# - ADDED ENTRY--CORPORATE NAME | |
| Corporate name or jurisdiction name as entry element | SpringerLink (Online service) |
| 773 0# - HOST ITEM ENTRY | |
| Title | Springer eBooks |
| 776 08 - ADDITIONAL PHYSICAL FORM ENTRY | |
| Display text | Printed edition: |
| International Standard Book Number | 9783642139314 |
| 830 #0 - SERIES ADDED ENTRY--UNIFORM TITLE | |
| Uniform title | Studies in Computational Intelligence, |
| -- | 1860-949X ; |
| Volume number/sequential designation | 291 |
| 856 40 - ELECTRONIC LOCATION AND ACCESS | |
| Uniform Resource Identifier | http://dx.doi.org/10.1007/978-3-642-13932-1 |
| 912 ## - | |
| -- | ZDB-2-ENG |
No items available.