Advances in reinforcement learning and their implications for intelligent control
作者： S.D. Whitehead , R.S. Sutton , D.H. Ballard
关键词: Intelligent control 、 Intelligent sensor 、 Adaptive control 、 Computer science 、 Intelligent decision support system 、 Robot learning 、 Reinforcement learning 、 Artificial intelligence 、 Active perception 、 Control system
摘要: The focus of this work is on control architectures that are based reinforcement learning. A number recent advances have contributed to the viability learning approaches intelligent surveyed. These include formalization relationship between and dynamic programming, use internal predictive models improve rate, integration with active perception. On basis these other results, it concluded base now in a position satisfy many criteria associated control. >
sci-hub.se 参考文章(21)
Richard C. Yee, Andrew G. Barto, Paul E. Utgoff, Sharad Saxena, Explaining temporal differences to create useful concepts for evaluating states national conference on artificial intelligence. pp. 882- 888 ,(1990)
Shimon Ullman, Visual routines Image understanding 1985-86. pp. 286- 344 ,(1987)
Philip E. Agre, David Chapman, Pengi: an implementation of a theory of activity national conference on artificial intelligence. pp. 268- 272 ,(1987)
Jeffrey C. Schlimmer, Ming Tan, Two case studies in cost-sensitive concept acquisition national conference on artificial intelligence. pp. 854- 860 ,(1990)
Richard S. Sutton, Integrated architecture for learning, planning, and reacting based on approximating dynamic programming international conference on machine learning. pp. 216- 224 ,(1990) , 10.1016/B978-1-55860-141-3.50030-4
Steven D. Whitehead, Dana H. Ballard, Active perception and reinforcement learning international conference on machine learning. pp. 179- 188 ,(1990) , 10.1016/B978-1-55860-141-3.50025-0
Long-Ji Lin, Self-improving reactive agents: case studies of reinforcement learning frameworks simulation of adaptive behavior. pp. 297- 305 ,(1991)
John J. Grefenstette, Incremental learning of control strategies with genetic algorithms international conference on machine learning. pp. 340- 344 ,(1989) , 10.1016/B978-1-55860-036-2.50087-4
Steven D. Whitehead, Dana H. Ballard, A role for anticipation in reactive systems that learn international conference on machine learning. pp. 354- 357 ,(1989) , 10.1016/B978-1-55860-036-2.50090-4
Philip E. Agre, The Dynamic Structure of Everyday Life Massachusetts Institute of Technology. ,(1988)