Research Interests

I recently started as a postdoc in the Departments of Psychology and Computer Science at Princeton University, working with Prof. Tom Griffiths. Updated website coming up soon!

I earned my PhD in Physics from Harvard in late 2019. My research is at the intersection of machine learning and computational cognitive science, both on a) leveraging methods from machine learning to shed light on process-level accounts of how humans reason, and b) using these insights to understand current black-box machine learning algorithms, as well as build better, more human-like artificial intelligence.

During my PhD, I worked with Sam Gershman in the Computational Cognitive Neuroscience Lab  at Harvard. My thesis is on trying to understand how humans infer probabilities in the real world. Specifically, how people might be trading-off the precision/accuracy and the computational costs of various algorithms for statistical inference. My focus is on how humans make best use of their limited resources, by utilizing structure in the environment to cheaply approximate otherwise expensive computations (i.e. by being "ecologically rational"). I study this by building computational models that make predictions about human behavior in various tasks, and subsequently testing these predictions with online behavioral experiments. I also worked on how the lens of ecological rationality provides new ways to both understand current machine learning systems (via diagnostic test sets) as well as improve them (via augmentation of their training environments).

General research interests: Cognitive psychology, computational cognitive modeling, stochastic approximations, amortized inference, bounded rationality, machine learning, artificial intelligence, Bayesian statistics, causal inference.

My publications page here is not up to date, please refer to my google scholar profile for an updated list!