A Genetic Operator for the Two Dimensional Stochastic Learning Cellular Automata

To construct the model of reinforcement learning systems ， we present 巴 da theoretic mod 巴 l of stochastic learning cellular automata (SLCA) in our previous pape r. Th e SLCA is an extended model of traditional cellular automaton ， defined as a stochastic c 巴 llularautomaton with its random environment There are three rule spaces for the SLC A: parallel ， sequential and mixture. Th is paper suggests a parallel SLCA with a genetic op 巴 ratorand applies it to the combinatorial optimi zation problems. Th巴 comput 巴 rsimulations of graph partition problem show that the convergence of SLCA is better than parallel mean fi 巴 ldalgorithm.