Degree
Bachelor of Science (B.S.)
Major(s)
Computer Science
Document Type
Immediate Open Access
Abstract
Artificial neural networks, which mimic the human brain's ability to learn from experiences, are increasingly being used to analyze complex datasets. However, proper structural configuration requires intuition, trial-and-error, and frequent human attention. This study investigates an automated alternative that uses a Darwinian evolutionary strategy to optimize the structure of a small-scale Modified National Institute of Standards and Technology (MNIST) image classification network. Using accuracy as a measure of fitness, it examines the effect of varying the amount each network learns prior to differential reproduction. The accuracy of optimized networks was significantly higher than random initial networks, being increased by up to 0.133% (1.435 standard deviations above initial mean accuracy). The use of evolutionary strategies in design holds promise for producing networks that are appreciably more accurate than randomly-generated networks without a large ongoing input of human attention.
DOI
https://doi.org/10.54718/WGWF9941
Date Defended
5-1-2018
Thesis Director
Archibald, Christopher
Second Committee Member
Bethel, Cindy L.
Third Committee Member
Gardner, Becky
Recommended Citation
Dinep-Schneider, Nicholas, "On the feasibility of using genetic algorithms to optimize the structure of small multilayer perceptrons" (2018). Honors Theses. 31.
https://scholarsjunction.msstate.edu/honorstheses/31
