In this paper, we propose a genetic network programming (GNP) architecture using a coevolution model called automatically defined groups (ADG). The GNP evolves networks for describ...
Partially Observable Markov Decision Processes (POMDPs) provide a general framework for AI planning, but they lack the structure for representing real world planning problems in a...
Background: We consider the problem of parameter estimation (model calibration) in nonlinear dynamic models of biological systems. Due to the frequent ill-conditioning and multi-m...
Maria Rodriguez-Fernandez, Jose A. Egea, Julio R. ...
Mixture-of-Experts (MoE) systems solve intricate problems by combining results generated independently by multiple computational models (the “experts”). Given an instance of a...
Genetic algorithms have been successfully applied to many difficult problems but there have been some disappointing results as well. In these cases the choice of the internal repre...