A-Brain: the multiple problems solver

Abstract

Intelligence is strongly related to the ability of solving different problems by a single system. General problems solvers such as Artificial Neural Networks, Evolutionary Algorithms, Particle Swarm etc, have traditionally been tested against one problem at one time. The purpose of this research is to build a complex and adaptive system able to solve multiple (and different) problems. The proposed system, called A-Brain, consists of several connected components (a Decision Maker, a Trainer and several Problem Solvers) which provide a base for building complex problem solvers. The A-Brain system is applied for solving some well-known problems in the field of symbolic regression. Numerical experiments show that A-Brain system is able to perform very well on the considered test problems.