AN EVOLUTIONARY ALGORITHM THAT CONSTRUCTS RECURRENT NEURAL NETWORKS

Standard methods for simultaneously inducing the structure and weights of recurrent neural networks limit every task to an assumed class of architectures. Such a simplification is necessary since the interactio ns between network structure and function are not well understood. Evo lutionary computations, which include genetic algorithms and evolution ary programming, are population-based search methods that have shown p romise in many similarly complex tasks. This paper argues that genetic algorithms are inappropriate for network acquisition and describes an evolutionary program, called GNARL, that simultaneously acquires both the structure and weights for recurrent networks. GNARL's empirical a cquisition method allows for the emergence of complex behaviors and to pologies that are potentially excluded by the artificial architectural constraints imposed in standard network induction methods.