neural networks research group
areas
people
projects
demos
publications
software/data
Speciation in NEAT (2010)
Timothy Nodine
This paper investigates speciation in NEAT and its effect on NEAT's performance. Speciation is the set of processes by which NEAT creates, maintains and uses several disjoint groups of similar genomes for guiding reproduction. This paper also examines the issue of adoption, where mating within a species produces an individual in a different species. It then examines the relationship between NEAT's performance, the adoption rate and the number of species. The paper demonstrates that while species and adoptions are necessary, too many species and adoptions harm the algorithm's performance.
View:
PDF
Citation:
Technical Report HR-10-06, Department of Computer Science, The University of Texas at Austin, 2010.
Bibtex:
@techreport{nodine:ugthesis10, title={Speciation in NEAT}, author={Timothy Nodine}, number={HR-10-06}, institution={Department of Computer Science, The University of Texas at Austin}, type={Undergraduate Honors Thesis}, url="http://nn.cs.utexas.edu/?nodine:ugthesis10", year={2010} }
People
Timothy Nodine
Undergraduate Alumni
Projects
Learning Strategic Behavior in Sequential Decision Tasks
2009 - 2014
Areas of Interest
Evolutionary Computation
Neuroevolution