Mobile Particle Swarm Optimization on Manifolds

Particle swarm optimization (PSO) is an increasingly used optimization technique which is derived from biology. Generally, all particles are assumed in a linear space that is uniformly distributed. A manifold is nearly a curved space. The purpose of this paper is to present a mobile particle swarm optimization framework on manifolds, that is, on curved space. The newly proposed MPSO will apply for nonlinear, non-uniform data distribution. Convergence analysis is presented and performance of the new algorithm is examined with experimental data given.