Genetic crossover in the evolution of time-dependent neural networks

Jason Orlosky, Tim Grabowski

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Neural networks with temporal characteristics such as asynchronous spiking have made much progress towards biologically plausible artificial intelligence. However, genetic approaches for evolving network structures in this area are still relatively unexplored. In this paper, we examine a specific variant of time-dependent spiking neural networks (NN) in which the spatial and temporal relationships between neurons affect output. First, we built and customized a standard NN implementation to more closely model the time-delay characteristics of biological neurons. Next, we tested this with simulated tasks such as food foraging and image recognition, demonstrating success in multiple domains. We then developed a genetic representation for the network that allows for both scalable network size and compatibility with genetic crossover operations. Finally, we analyzed the effects of genetic crossover algorithms compared to random mutations on the food foraging task. Results showed that crossover operations based on node usage converge on a local maximum more quickly than random mutations, but suffer from genetic defects that reduce overall population performance.

Original languageEnglish (US)
Title of host publicationGECCO 2021 - Proceedings of the 2021 Genetic and Evolutionary Computation Conference
PublisherAssociation for Computing Machinery, Inc
Pages885-891
Number of pages7
ISBN (Electronic)9781450383509
DOIs
StatePublished - Jun 26 2021
Externally publishedYes
Event2021 Genetic and Evolutionary Computation Conference, GECCO 2021 - Virtual, Online, France
Duration: Jul 10 2021Jul 14 2021

Publication series

NameGECCO 2021 - Proceedings of the 2021 Genetic and Evolutionary Computation Conference

Conference

Conference2021 Genetic and Evolutionary Computation Conference, GECCO 2021
Country/TerritoryFrance
CityVirtual, Online
Period7/10/217/14/21

Keywords

  • Evolutionary computing
  • Genetic crossover
  • Simulation
  • Spiking neural network

ASJC Scopus subject areas

  • Genetics
  • Computational Mathematics

Fingerprint

Dive into the research topics of 'Genetic crossover in the evolution of time-dependent neural networks'. Together they form a unique fingerprint.

Cite this