My research this summer has had 2 focii. The first is examining bias in Large Language Models through text based games. I am primarily working with Professor Hartvigsen on this.

The second avenue is focused on developing and improving advanced reinforcement learning algorithms for the healthcare space. We are devloping a modified DQN / Critor Actor network to tackle the combinatoric action space that is modern healthcare. We are currently using a modified gridworld as a proxy for healthcare data. This project is led by Matt Landers and Professor Hartvigsen.

Slides

A list of the materials that I created can be found here.