Davin Lin's Final Project Presentation (SEE-Classify)

In the summer of 2021 Davin Lin worked with me as part of the ICER/ACERS REu program on his project titled "Simple Evolutionary Exploration in Classification Algorithms for Supervised Learning"

Genetic algorithms (GAs) find good solutions to search problems through a process inspired by evolution. Solutions are randomly selected and tested using a fitness function. The best solutions undergo changes (mutations) over multiple iterations (generations) to try and find better solutions. There have been several studies that use GAs to search over hyper-parameters of machine learning algorithms to learn values that work well for specific problems. For example, one popular study performs genetic search over breast cancer data to find the best supervised learning classification algorithm [ref]. One of the main benefits of using genetic search is that it can be domain independent. Any scientist can use this method to find a well-performing classification algorithm for their dataset. This summer, we will leverage this property by extending an existing GA framework to adopt classifiers. This will allow scientists from any field to search a classifier “algorithm space” without having to recreate the machinery to do so. We will test our software and perform a proof-of-concept by attempting to partially reproduce the experimental results of previous works.

Here is a video from his final presentation: