Networks of Learning Automata: Techniques for Online Stochastic Optimization

Networks of Learning Automata: Techniques for Online Shastic Optimization is a comprehensive account of learning automata models with emphasis on multiautomata systems. It considers synthesis of complex learning structures from simple building blocks and uses shastic algorithms for refining probabilities of selecting actions. Mathematical analysis of the behavior of games and feedforward networks is provided. Algorithms considered here can be used for online optimization of systems based on noisy measurements of performance index. Also, algorithms that assure convergence to the global optimum are presented. Parallel operation of automata systems for improving speed of convergence is described. The authors also include extensive discussion of how learning automata solutions can be constructed in a variety of applications.

Thathachar and Sastry (both electrical engineering, Indian Institute of Science, Bangalore) consider the synthesis of complex learning structures from simple building blocks. The building block in this case is the single learning automaton which learns to select the best action by repeated interactions with an unknown random environment, which the authors use to create systems consisting of several learning automata, such as games and feedforward networks. Mathematical analysis of their behavior is provided, along with discussion of the construction of learning automata solutions in a variety of applications. Suitable as a text in graduate level courses and as a reference work for researchers. Annotation ©2004 Book News, Inc., Portland, OR