An Adaptive Robot Motivational System
作者： George Konidaris , Andrew Barto
DOI: 10.1007/11840541_29
关键词:
摘要: We present a robot motivational system design framework The represents the underlying (possibly conflicting) goals of as set drives, while ensuring comparable drive levels and providing mechanism for priority adaptation during robot's lifetime resulting reward signals are compatible with existing reinforcement learning methods balancing multiple functions illustrate an experiment that demonstrates some its benefits.
springer.com
core.ac.uk
springerlink.com参考文章(20)
Dana Ballard, Nathan Sprague, Multiple-goal reinforcement learning with modular Sarsa(O) international joint conference on artificial intelligence. pp. 1445- 1447 ,(2003)
Stewart W. Wilson, Jean-Arcady Meyer, From Animals to Animats: Proceedings of The First International Conference on Simulation of Adaptive Behavior (Complex Adaptive Systems) ,(1990)
Peter M. Todd, Bruce M. Blumberg, No Bad Dogs: Ethological Lessons for Learning in Hamsterdam* ,(1999)
Jean-Arcady Meyer, Stewart W Wilson, ECOLE NORMALE SUPERIEURE PARIS (FRANCE) GROUPE DE BIOINFORMATIQUE, From Animals to Animats: Proceedings of the International Conference on Simulation of Adaptive Behavior (1st) Held in Paris, France on 24-28 September 1990 ,(1991)
Kathleen O'Connor, K O'Connor, Learning: An Introduction ,(1968)
Mark Humphrys, Action Selection methods using Reinforcement Learning University of Cambridge. ,(1996)
Damien Ernst, Arthur Louette, Introduction to Reinforcement Learning MIT Press. ,(1998)
Thomas Bösser, David McFarland, Intelligent behavior in animals and robots ,(1993)
Hugues Bersini, Reinforcement learning for homeostatic endogenous variables simulation of adaptive behavior. pp. 325- 333 ,(1994)
Steven Whitehead, Jonas Karlsson, Josh Tenenberg, Learning Multiple Goal Behavior via Task Decomposition and Dynamic Policy Merging Robot Learning. ,vol. 3, pp. 45- 78 ,(1993) , 10.1007/978-1-4615-3184-5_3