Title :
Associative reinforcement learning of real-valued functions
Author :
Gullapalli, VijayKumar
Author_Institution :
Dept. of Comput. & Inf. Sci., Massachusetts Univ., Amherst, MA, USA
Abstract :
The author describes an algorithm, called the stochastic real-valued (SRV) algorithm, that uses evaluative performance feedback to learn associative maps from input vectors to real-valued actions. This algorithm is based on the pioneering work of A.G. Barto and P. Anandan (1985), in synthesizing associative reinforcement learning (ARL) algorithms using techniques from pattern classification and automata theory. A strong convergence theorem is presented that implies a form of optimal performance under certain general conditions of the SRV algorithm on ARL tasks. Simulation results are presented to illustrate the convergence behavior of the algorithm under the conditions of the theorem. The robustness of the algorithm is also demonstrated by simulations in which some of the conditions of the theorem are violated
Keywords :
automata theory; convergence; learning systems; neural nets; pattern recognition; associative reinforcement learning; automata theory; evaluative performance feedback; neural nets; pattern classification; pattern recognition; stochastic real-valued algorithm; strong convergence theorem; Algorithm design and analysis; Convergence; Learning automata; Learning systems; Probability distribution; Robustness; Stochastic processes; Stochastic systems; Supervised learning; Uncertainty;
Conference_Titel :
Systems, Man, and Cybernetics, 1991. 'Decision Aiding for Complex Systems, Conference Proceedings., 1991 IEEE International Conference on
Conference_Location :
Charlottesville, VA
Print_ISBN :
0-7803-0233-8
DOI :
10.1109/ICSMC.1991.169893