| Because DNA encodes complexity on a massive scale, interest has grown in recent years in the powerful mapping between genotype and phenotype in nature. An important question for evolutionary computation is whether an artificial encoding of similar capacity can be devised to evolve highly complex artificial structures on computers. Because DNA maps to the phenotype through a process of development, a key challenge in designing such an encoding is to identify the right level of algorithmic abstraction of the development of the embryo in nature. This talk will argue that it is possible to capture the essential properties of developmental encoding at a much higher level of abstraction than previously thought. This new developmentally-motivated encoding, called Compositional Pattern Producing Networks (CPPNs), can represent and evolve complex spatial patterns with recognizable natural regularities. CPPNs are further extended to represent large-scale neural networks by exploiting a surprisingly simple geometric trick. The result is a novel algorithm called HyperNEAT (Hypercube-based NeuroEvolution of Augmenting Topologies) that can artificially evolve working neural networks with millions of connections that exploit geometric regularities in several problem domains. |