Background

My research interests lie at the intersection of microeconomic theory, algorithmic logic, and finance theory. While the areas of application are in economics and finance, my work often uses ideas from theoretical computer science and mathematical logic. 

The perspective that I like to take is that logicians following Frege, alongside neuroscientists following McCulloch and Pitts, developed extremely nuanced accounts of individual reasoning, while largely leaving out collective interaction — whereas economists built careful descriptions of collective equilibrium interaction, while largely glossing over non-equilibrium, individual reasoning. To unlock deeper insights, the next step is to scale the neuroscientists' and logicians' individual cognitive models into economically relevant models of collective interaction. Crucially, the reasoning agent at the center could be naturally or artificially intelligent: the economic theory should be invariant to this distinction.

This is a shift from the given tradition in economics: instead of taking an equilibrium notion as the primitive of an economic interaction, this perspective advocates taking the algorithm or logic which captures an individual agent's reasoning (equilibrium or not) as the primitive. To give an example, a broad question that I've puzzled over for some time is whether there is a notion of gameplay that is invariant to whether an agent is using gradient-descent-based reasoning or regular Bayesian reasoning. In Finance, my work explores how market rules can affect market outcomes, and the tradeoffs that agents have to make in complex interactive landscapes.