Sustaining behavioral diversity in NEAT

Authors:
Hirotaka Moriguchi;Shinichi Honiden
Affiliations:
The University of Tokyo, Tokyo, Japan;The University of Tokyo, Tokyo, Japan
Venue:
Proceedings of the 12th annual conference on Genetic and evolutionary computation
Year:
2010

Citing 11
Cited 2

Genetic algorithms with sharing for multimodal function optimization

Proceedings of the Second International Conference on Genetic Algorithms on Genetic algorithms and their application
Evolving neural networks through augmenting topologies

Evolutionary Computation
A hypercube-based encoding for evolving large-scale neural networks

Artificial Life
An empirical analysis of value function-based and policy search reinforcement learning

Proceedings of The 8th International Conference on Autonomous Agents and Multiagent Systems - Volume 2
Sustaining diversity using behavioral information distance

Proceedings of the 11th Annual conference on Genetic and evolutionary computation
How novelty search escapes the deceptive trap of learning to learn

Proceedings of the 11th Annual conference on Genetic and evolutionary computation
Using behavioral exploration objectives to solve deceptive problems in neuro-evolution

Proceedings of the 11th Annual conference on Genetic and evolutionary computation
Discovering several robot behaviors through speciation

Evo'08 Proceedings of the 2008 conference on Applications of evolutionary computing
Critical factors in the empirical performance of temporal difference and evolutionary methods for reinforcement learning

Autonomous Agents and Multi-Agent Systems
Fitness sharing and niching methods revisited

IEEE Transactions on Evolutionary Computation
The similarity metric

IEEE Transactions on Information Theory

Encouraging behavioral diversity in evolutionary robotics: An empirical study

Evolutionary Computation
Evolving team behaviors with specialization

Genetic Programming and Evolvable Machines

Quantified Score

Hi-index	0.00

Visualization

Abstract

Niching schemes, which sustains population diversity and let an evolutionary population avoid premature convergence, have been extensively studied in the research field of evolutionary algorithms. Neuroevolutionary (NE) algorithms, such as NEAT, have also benefitted from niching. However, the latest research indicates that the use of genotype- or phenotype-similarity-based niching schemes in NE algorithms is not highly effective because these schemes have difficulty sustaining the behavioral diversity in the environment. In this paper, we propose a novel niching scheme that takes into consideration both the phenotypic and behavioral diversity, and then integrate it with NEAT. An experimental analysis revealed that the proposed algorithm outperforms the original NEAT for various problem settings. More interestingly, it performs especially well for problems with a high noise level and large state space. Since these features are common in problems to which NEAT is applied, the proposed algorithm should be effective in practice.