On the alleviation of the problem of local minima in back-propagation
Proceedings of second world congress on Nonlinear analysts
Recent approaches to global optimization problems through Particle Swarm Optimization
Natural Computing: an international journal
Hi-index | 0.01 |
This paper considers the application of a novel optimization method, namely Particle Swarm Optimization, to compute Nash equilibria. The problem of computing equilibria is formed as one of detecting the global minimizers of a real-valued, non-negative, function. To detect more than one global minimizers of the function at a single run of the algorithm and address effectively the problem of local minima, the recently proposed Deflection technique is employed. The performance of the proposed algorithm is compared to that of algorithms implemented in the popular game theory software suite, GAMBIT. Conclusions are derived.