TSMC
2002
14 years 4 months ago
2002
An extended algorithm of the relative reward strength algorithm is proposed. It is shown that the proposed algorithm ensures the convergence with probability 1 to the optimal path ...