Barrett

Barrett, The Honors College at Arizona State University proudly showcases the work of undergraduate honors students by sharing this collection exclusively with the ASU community.

Barrett accepts high performing, academically engaged undergraduate students and works with them in collaboration with all of the other academic units at Arizona State University. All Barrett students complete a thesis or creative project which is an opportunity to explore an intellectual interest and produce an original piece of scholarly research. The thesis or creative project is supervised and defended in front of a faculty committee. Students are able to engage with professors who are nationally recognized in their fields and committed to working with honors students. Completing a Barrett thesis or creative project is an opportunity for undergraduate honors students to contribute to the ASU academic community in a meaningful way.

Displaying 1 - 1 of 1
Filtering by

Clear all filters

131424-Thumbnail Image.png

Bottom-Up Strategy Development: Evolving Metagames of Neural Network-Based Strategies

Description
In this paper, I describe the development of a unique approach to developing strategies for games in which success can only be measured by the final outcome of the game, preventing the use of heuristics. I created and evaluated evolutionary algorithms, applying them to develop strategies for tic-tac-toe. Strategies are…

In this paper, I describe the development of a unique approach to developing strategies for games in which success can only be measured by the final outcome of the game, preventing the use of heuristics. I created and evaluated evolutionary algorithms, applying them to develop strategies for tic-tac-toe. Strategies are comprised of neural networks with randomly initiated weights. A population of candidate strategies are created, each strategy competes individually against each other strategy, and evolutionary operators are applied to create subsequent generations of strategies. The set of strategies within a generation of the evolutionary algorithm forms a metagame that evolves as the algorithm progresses. Hypothesis testing shows that strategies produced by this approach significantly outperform a baseline of entirely random action, although they are still far from optimal gameplay.
ContributorsRodriguez, Julien Guillermo (Author) / Martin, Thomas (Thesis director) / Powers, Brian (Committee member) / College of Integrative Sciences and Arts (Contributor, Contributor) / Department of Information Systems (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05