An Evolutionary Autonomous Agent with Visual Cortex and Recurrent Spiking Columnar Neural Network

Rich Drewes¹, James Maciokas¹, Sushil J. Louis², and Philip Goodman¹

¹Brain Computation Laboratory
http://brain.cs.unr.edu

²Evolutionary Computing Systems Lab, University of Nevada, Reno NV 89557, USA
http://ecsl.cs.unr.edu

Abstract. Spiking neural networks are computationally more powerful than conventional artificial neural networks [1]. Although this fact should make them especially desirable for use in evolutionary autonomous agent research, several factors have limited their application. This work demonstrates an evolutionary agent with a sizeable recurrent spiking neural network containing a biologically motivated columnar visual cortex. This model is instantiated in spiking neural network simulation software and challenged with a dynamic image recognition and memory task. We use a genetic algorithm to evolve generations of this brain model that instinctively perform progressively better on the task. This early work builds a foundation for determining which features of biological neural networks are important for evolving capable dynamic cognitive agents.

LNCS 3102, p. 257 f.

Full article in PDF

lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2004