My research interests lie in the intersection of Natural Language Processing, Reinforcement Learning and Creativity. I am above all interested in building assistive, agentic tools to help humans in a wide range of human creative tasks. Creative tasks, I define, are tasks where reward functions are complex, unobserved, subjective or individual. Inspired by cognitive science, I am especially interested in how humans learn rewards through partial observation: for example, you or I can learn a lot about the process a scientist follows, simply by reading her articles. In other words, by observing just the end-state of the article generation process, we can make reasonable inferences about the entire trajectory, which can help us then model the rewards driving this process. I am pursuing projects in this direction in a number of different fields: computational journalism, AI4Science, music and public policy. If you are interested in any of these directions, please reach out!
Education
Postdoctoral Researcher, Stanford University
Ph.D. in Computer Science, University of Southern California
M.S. in Journalism, Columbia University Graduate School of Journalism
M.S. in Data Science, Columbia University Fu School of Engineering
B.S. in Computer Science, Columbia University
B.S. in Neuroscience, Columbia University
