Hierarchical discretized pursuit nonlinear learning automata
Athanasios V. Vasilakos, Georgios I. Papadimitriou, Constantinos T. Paximadis · 2002
A new absorbing multiaction learning automaton is presented which is epsilon-optimal. The proposed automaton (named HDPNRI) is a hierarchical discretized nonlinear one which utilizes a new pursuit learning algorithm. The HDPNRI automaton has the best performance (speed of convergence, CPU time and accuracy) among all the absorbing learning automata reported in the literature. Extensive simulation results indicate the superiority of HDPNRI's performance. Furthermore, it is proved that the HDPNRI learning automaton is epsilon-optimal in every stationary environment.>