Caleb Schultz Kisby
PhD Candidate, he/him
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.
Note: I’m on the research job market this year (Fall 2024). I am primarily looking for postdoc positions, but I am open to any opportunity to continue research in academia or industry.
Contact: cckisby (✨) gmail.com
Research
I study the foundations of machine learning and cognition from the perspective of logic. I’m especially interested in issues at the intersection of neural network learning, belief revision, dynamic epistemic logic, and descriptive complexity.
My current (thesis) work is on the theory of neuro-symbolic AI, where I use insights from logic to better understand and control the behavior of neural networks as they learn over time. My long-term goal is to bring these ideas to bear on two of the most critical issues in artifical intelligence: neural network verification and the possibility of AI alignment.