NeuroEvolution of Augmenting Topologies (NEAT) is a genetic algorithm for the generation of evolving artificial neural networks (a neuroevolution technique) developed by Ken Stanley in 2002 while at The University of Texas at Austin. It alters both the weighting parameters and structures of networks, attempting to find a balance between the fitness of evolved solutions and their diversity. It is based on applying three key techniques: tracking genes with history markers to allow crossover among topologies, applying speciation (the evolution of species) to preserve innovations, and developing topologies incrementally from simple initial structures ("complexifying").
Read more about Neuroevolution Of Augmenting Topologies: Performance, Complexification, Implementation
Famous quotes containing the word augmenting:
“The true thrift is always to spend on the higher plane; to invest and invest, with keener avarice, that he may spend in spiritual creation, and not in augmenting animal existence. Nor is the man enriched, in repeating the old experiments of animal sensation; nor unless through new powers and ascending pleasures he knows himself by the actual experience of higher good to be already on the way to the highest.”
—Ralph Waldo Emerson (1803–1882)