My Research

Mission Statement

My research is in the design of efficient algorithms that expand the reliability and efficacy of multi-agent learning systems by grounding them in reinforcement learning and game theoretic principles.



Research Interests


  • Multi-agent Reinforcement Learning
  • Game theory, game-solving in large games, equilibrium selection
  • Game-solving sample-efficiency

Where I Think the Field is Going


  • Incorporation of offline or model-based methods for game-solving
  • Language models and human feedback for strategy generation
  • Incorporating bounded rationality for agent-based modeling


Have Fun Looking Around

Just like research, I tried to make this page (hopefully) appealing and fun. I'm not saying my job is happy and dandy all the time. I just think that "serious" things deserve love sometimes, especially things we care about.

Current Project

Currently working on two projects. One will (hopefully) be published soon and the other is still half-baked. Will post it soon.

Later Class Projects

Fully Online Decision Transformer

A good friend of mine and I built upon the recent Decision Transformer (DT) for an NLP class at UMich. We adapted DT to online contexts, training DT tabula rasa with a SOTA online RL algorithm for exploration.

RL for Adversarial Text Generation

A group of friends and I gave our take on adversarial text generation. We used an RL method that adversarially swapped words in input texts to fool state-of-the-art text classification models.

GNN's for Register Allocation

Another group of friends and I applied machine learning to a compiler problem: register allocation. If the problem can be reduced to a graph coloring problem, why not apply machine learning?

Ant-Inspired Decentralized MARL

Undergraduate Honors Thesis

This project created RL agents that imitate how ants communicate through pheromones to induce a desired joint behavior: decentralized coordination through indirect communication. Agents learned how to modify the environment for traversal through single-agent RL and a coordination algorithm to have multiple agents cooperatively navigate to a goal location.

What is Stigmergy?

A mechanism of indirect coordination through the environment, characterized by traces left in the environment by agents to stimulate the performance of succeeding actions by the same or different agents.