Final project for COS597C — Theory of Natural Algorithms (Fall 2024) at Princeton University. View the paper at report.pdf.
There's 3 parts to the paper. First, I overview the work done in Ohtsuki et al. (2006) to introduce evolutionary dynamics on graphs from a game-theoretic perspective. This overview also provides an analytical solution to a game, but for my original work, I focus on numerical computer simulations. Second, I do a quick literature review of related works (mostly done by Ohtsuki and/or Nowak) and discuss the implications of these works. Finally, I present my original work on including sleep into the evolutionary dynamics. I show that sleep can be a dominant strategy in some contexts despite it having drawbacks for an individual. Additionally, I provide suggestions for future work, especially with regards to bridging the gap between experimental data-first analysis and mathematical model-first theory.
This report is in no way comprehensive, but eventually, I might try to expand on it as a full paper on theoretical sleep dynamics.
The individual simulation units are in the modules directory. To run the simulations, modify main.ts to include the desired parameters and then run deno run main.ts in the root directory. This will output the results to the console. It will most likely work with NodeJS as well, but I have not tested it.
This project is licensed under the BSD 3-Clause License. See the LICENSE file for more information.