Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


RMTSignatures.pdf (795.15 KB)

ETD Abstract Container

Abstract Header

Evolving Neural Networks Through Random Augmentation and Sexual Reproduction

Robinson, Andrew Locke
http://rave.ohiolink.edu/etdc/view?acc_num=akron1606912380236988

Abstract Details

2020, Master of Science, University of Akron, Computer Science.
Neuroevolution methods which evolve the topology of a neural network as well as their weights are poised to be powerful tools in machine learning as they can potentially adapt to any degree of complexity. The neuroevolution method presented in this paper, Random Augmentation and Sexual Reproduction (RASR), outperforms all existing fixed-topology and augmenting topology neuroevolution methods on a difficult reinforcement learning benchmark. RASR achieves this result by implementing a novel reproduction technique which performs extensive crossover and allows the combination of networks of varying topologies. RASR also implements a novel augmentation process which promotes a stable complexification through generations without the need for speciation. Further, all the resultant networks produced by RASR started from the most minimal structure, a network without any connections and only input and output nodes. This shows that RASR can create complex solutions not only quickly but also without the bias of starting from a promising architecture.
Zhong-Hui Duan (Advisor)
Michael Collard (Committee Member)
Timothy O'Neil (Committee Member)
59 p.
Computer Science
neuroevolution; artificial intelligence; neural networks; genetic algorithms; augmenting topology;

Recommended Citations

Citations

  • Robinson, A. L. (2020). Evolving Neural Networks Through Random Augmentation and Sexual Reproduction [Master's thesis, University of Akron]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=akron1606912380236988

    APA Style (7th edition)

  • Robinson, Andrew. Evolving Neural Networks Through Random Augmentation and Sexual Reproduction. 2020. University of Akron, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=akron1606912380236988.

    MLA Style (8th edition)

  • Robinson, Andrew. "Evolving Neural Networks Through Random Augmentation and Sexual Reproduction." Master's thesis, University of Akron, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=akron1606912380236988

    Chicago Manual of Style (17th edition)

akron1606912380236988
168
© 2020, all rights reserved.
This open access ETD is published by University of Akron and OhioLINK.