Evolution of Neural Networks Through Incremental Acquisition of Neural Structures
In this contribution we present a novel method, called Evolutionary Acquisition of Neural Topologies (EANT), of evolving the structures and weights of neural networks. The method introduces an efficient and compact genetic encoding of a neural network onto a linear genome that enables one to evaluate the network without decoding it. The method uses a meta-level evolutionary process where new structures are explored at larger time-scale and the existing structures are exploited at lower time-scale. This enables it to find minimal neural structures for solving a given learning task.