TSMC
2002
13 years 11 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 ...