I am a senior studying EECS at UC Berkeley. I am interested in multi-agent reinforcement learning, which I work on at BAIR.
Working on deep reinforcement learning algorithms applied to multiagent games. Exploring techniques to make agents less exploitable by adapting quickly to new opponents.
Integrated Gmail Ads data into a MapReduce pipeline that generates terabytes of clean training data by combining data from multiple sources. Helped with the development of several internal tools, including a Gmail Ads serving stack diagnostic tool, a user model viewer and editor, and a visual user event inspector.
Worked with Python frameworks such as PyQt and OpenGL to develop tools for computer fracking simulations.
In high school, I went to summer camps for game design and artificial intelligence. When I watched the historic Alpha Go vs Lee Sedol, I knew that I wanted to work on reinforcement learning. In my research, I try to combine my passion for games and AI, and my goal is to create agents that can reason about strategy and their opponents. One of the best feelings is when the agents teach us something new about a game that no human has figured out.