Enhancing es-hyperneat to evolve more complex regular neural networks

Authors:
Sebastian Risi;Kenneth O. Stanley
Affiliations:
University of Central Florida, Orlando, FL, USA;University of Central Florida, Orlando, FL, USA
Venue:
Proceedings of the 13th annual conference on Genetic and evolutionary computation
Year:
2011

Citing 13
Cited 2

Evolving neural networks through augmenting topologies

Evolutionary Computation
Creating high-level components with a generative representation for body-brain evolution

Artificial Life
A Taxonomy for artificial embryogeny

Artificial Life
Evolving modular genetic regulatory networks

CEC '02 Proceedings of the Evolutionary Computation on 2002. CEC '02. Proceedings of the 2002 Congress - Volume 02
Compositional pattern producing networks: A novel abstraction of development

Genetic Programming and Evolvable Machines
Picbreeder: evolving pictures collaboratively online

Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
A hypercube-based encoding for evolving large-scale neural networks

Artificial Life
A case study on the critical role of geometric regularity in machine learning

AAAI'08 Proceedings of the 23rd national conference on Artificial intelligence - Volume 2
Competitive coevolution through evolutionary complexification

Journal of Artificial Intelligence Research
Evolving coordinated quadruped gaits with the HyperNEAT generative encoding

CEC'09 Proceedings of the Eleventh conference on Congress on Evolutionary Computation
Autonomous evolution of topographic regularities in artificial neural networks

Neural Computation
Evolving the placement and density of neurons in the hyperneat substrate

Proceedings of the 12th annual conference on Genetic and evolutionary computation
Investigating whether hyperNEAT produces modular neural networks

Proceedings of the 12th annual conference on Genetic and evolutionary computation

Evolving neural networks

Proceedings of the 14th annual conference companion on Genetic and evolutionary computation
An enhanced hypercube-based encoding for evolving the placement, density, and connectivity of neurons

Artificial Life

Quantified Score

Hi-index	0.00

Visualization

Abstract

The recently-introduced evolvable-substrate HyperNEAT algorithm (ES-HyperNEAT) demonstrated that the placement and density of hidden nodes in an artificial neural network can be determined based on implicit information in an infinite-resolution pattern of weights, thereby avoiding the need to evolve explicit placement. However, ES-HyperNEAT is computationally expensive because it must search the entire hypercube, and was shown only to match the performance of the original HyperNEAT in a simple benchmark problem. Iterated ES-HyperNEAT, introduced in this paper, helps to reduce computational costs by focusing the search on a sequence of two-dimensional cross-sections of the hypercube and therefore makes possible searching the hypercube at a finer resolution. A series of experiments and an analysis of the evolved networks show for the first time that iterated ES-HyperNEAT not only matches but outperforms original HyperNEAT in more complex domains because ES-HyperNEAT can evolve networks with limited connectivity, elaborate on existing network structure, and compensate for movement of information within the hypercube.