TEXPLORE: Temporal Difference Reinforcement Learning for Robots and Time-Constrained Domains

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. O...

Πλήρης περιγραφή

Λεπτομέρειες βιβλιογραφικής εγγραφής
Κύριος συγγραφέας: Hester, Todd (Συγγραφέας)
Συγγραφή απο Οργανισμό/Αρχή: SpringerLink (Online service)
Μορφή: Ηλεκτρονική πηγή Ηλ. βιβλίο
Γλώσσα:English
Έκδοση: Heidelberg : Springer International Publishing : Imprint: Springer, 2013.
Σειρά:Studies in Computational Intelligence, 503
Θέματα:
Διαθέσιμο Online:Full Text via HEAL-Link
LEADER 03418nam a22004935i 4500
001 978-3-319-01168-4
003 DE-He213
005 20151103123158.0
007 cr nn 008mamaa
008 130623s2013 gw | s |||| 0|eng d
020 |a 9783319011684  |9 978-3-319-01168-4 
024 7 |a 10.1007/978-3-319-01168-4  |2 doi 
040 |d GrThAP 
050 4 |a Q342 
072 7 |a UYQ  |2 bicssc 
072 7 |a COM004000  |2 bisacsh 
082 0 4 |a 006.3  |2 23 
100 1 |a Hester, Todd.  |e author. 
245 1 0 |a TEXPLORE: Temporal Difference Reinforcement Learning for Robots and Time-Constrained Domains  |h [electronic resource] /  |c by Todd Hester. 
264 1 |a Heidelberg :  |b Springer International Publishing :  |b Imprint: Springer,  |c 2013. 
300 |a XIV, 165 p. 55 illus. in color.  |b online resource. 
336 |a text  |b txt  |2 rdacontent 
337 |a computer  |b c  |2 rdamedia 
338 |a online resource  |b cr  |2 rdacarrier 
347 |a text file  |b PDF  |2 rda 
490 1 |a Studies in Computational Intelligence,  |x 1860-949X ;  |v 503 
505 0 |a Introduction -- Background and Problem Specification -- Real Time Architecture -- The TEXPLORE Algorithm -- Empirical Evaluation -- Further Examination of Exploration -- Related Work -- Discussion and Conclusion -- TEXPLORE Pseudo-Code. 
520 |a 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. 
650 0 |a Engineering. 
650 0 |a Image processing. 
650 0 |a Computational intelligence. 
650 0 |a Robotics. 
650 0 |a Automation. 
650 1 4 |a Engineering. 
650 2 4 |a Computational Intelligence. 
650 2 4 |a Image Processing and Computer Vision. 
650 2 4 |a Robotics and Automation. 
710 2 |a SpringerLink (Online service) 
773 0 |t Springer eBooks 
776 0 8 |i Printed edition:  |z 9783319011677 
830 0 |a Studies in Computational Intelligence,  |x 1860-949X ;  |v 503 
856 4 0 |u http://dx.doi.org/10.1007/978-3-319-01168-4  |z Full Text via HEAL-Link 
912 |a ZDB-2-ENG 
950 |a Engineering (Springer-11647)