Lotto-type competitive learning with particle swarm features

This correspondence describes our attempts of incorporating particle swarm features into competitive learning. We first reinterpret some of the symbols and notations used in particle swarm optimisation (PSO) algorithms in the light of competitive learning. Three versions of modifications to the classical frequency-sensitive competitive learning are presented. Their strengths and weaknesses are highlighted. This then enables us to introduce particle swarm like features into our lotto type competitive learning. Experimental results indicate that, like the PSO algorithms, a careful selection of the values for the control parameters is necessary for the successful convergence of particles in some of the proposed algorithms. Two new algorithms, however, show better robustness towards the values of the control parameters.