DocumentCode :
446683
Title :
A reinforcement learning method based on adaptive simulated annealing
Author :
Atiya, Amir F. ; Parlos, Alexander G. ; Ingber, Lester
Author_Institution :
Dept. of Comput. Eng., Cairo Univ., Giza, Egypt
Volume :
1
fYear :
2003
fDate :
27-30 Dec. 2003
Firstpage :
121
Abstract :
Reinforcement learning is a hard problem and the majority of the existing algorithms suffer from poor convergence properties for difficult problems. In this paper we propose a new reinforcement learning method that utilizes the power of global optimization methods such as simulated annealing. Specifically, we use a particularly powerful version of simulated annealing called adaptive simulated annealing (ASA) (Ingber, 1989). Towards this end we consider a batch formulation for the reinforcement learning problem, unlike the online formulation almost always used. The advantage of the batch formulation is that it allows state-of-the-art optimization procedures to be employed, and thus can lead to further improvements in algorithmic convergence properties. The proposed algorithm is applied to a decision making test problem, and it is shown to obtain better results than the conventional Q-learning algorithm.
Keywords :
decision making; learning (artificial intelligence); simulated annealing; adaptive simulated annealing; global optimization; reinforcement learning; Aggregates; Computational modeling; Convergence; Dynamic programming; Learning; Mechanical engineering; Mechanical factors; Simulated annealing; Telephony; Testing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Circuits and Systems, 2003 IEEE 46th Midwest Symposium on
ISSN :
1548-3746
Print_ISBN :
0-7803-8294-3
Type :
conf
DOI :
10.1109/MWSCAS.2003.1562233
Filename :
1562233
Link To Document :
بازگشت