I am a Senior Researcher at Microsoft Research AI in the productivity and intelligence group. My research interests include causal modeling and inference, probabilistic programming, and algorithmic decision-making.
I work on using multiverse counterfactual reasoning to power new capabilities in AI. Humans reason this way with “would have, could have, should have” modal language such as “I took the role at MSR, and now I’m happy. Had I taken the other role, I would have been sad.” This reasoning involves a prediction of an outcome (level of happiness) in a parallel world (where I took another role) based on outcome data from this world (I took the MSR role and now I’m happy). A similar pattern in reinforcement learning would be an agent that thinks, “I did this action and got this much reward, how much reward would I have attained had I done a different action?”
Multiverse counterfactual reasoning is a challenge for machine learning because we don’t have training data with labels from parallel worlds. I leverage deep learning, probabilistic programming, and programmable inference to help data scientists and domain experts encode the inductive biases needed to make algorithmic multiverse counterfactual reasoning possible. My goal is to use these algorithms to create primitives for building agents with higher-order “system 2” cognitive abilities such as common sense, introspection, and theory of mind.
My work targets applications of AI that support human decision-making about complex systems. I have a particular interest in decision-making in economics, energy, and agriculture. My research focuses on developing technology that powers new Azure AI services in these and other verticals.
Before joining MSR, I worked as a research engineer writing production code for probabilistic decision-making under uncertainty. I received my Ph.D. in statistics from Purdue. I’m a Johns Hopkins SAIS alumni and a graduate of the Hopkins-Nanjing Center.
To learn more about what I’m working on, read my newsletter.