Caleb Schultz Kisby

PhD Candidate, he/him

me.jpg

Welcome! I’m a PhD Candidate in Computer Science at Indiana University, co-supervised by Larry Moss and Saúl Blanco. I’m a member of the IU logic group and plwonks.

Research

I study the foundations of machine learning and cognition, often through the lens of formal logic. My current work is in the theory of neuro-symbolic AI, through which I hope to (1) better understand the power and limitations of neural network learning, (2) bridge the gap between neural network learning and belief revision (dynamic epistemic updates), and (3) explore the possibility of neural network alignment.

More generally, I like to bring my logic training to bear on problems in cognitive science. I also have some background in formal verification, and enjoy verifying my proofs in Agda and Lean whenever I can.