Paul Krogmeier
I’m a postdoctoral fellow at Harvard advised by Nada Amin and Walter Fontana. My primary area of research touches programming languages, logic, and program synthesis. I received a PhD from the University of Illinois advised by Madhusudan Parthasarathy.
I’ll be joining the University of Colorado Boulder as an assistant professor in computer science starting Fall 2026 and am recruiting PhD students. If my research interests you then consider applying to the CS PhD program and mentioning my name in your application.
I’m also a jazz alto saxophonist and always interested in jamming with others 🎷.
Some Interests
- Interpretability and efficiency in machine learning
- Symbolic learning, program synthesis
- Descriptive complexity, finite model theory
- Reliable computation with unreliable components, fault tolerance
- Computational capabilities of neural networks
-
Musical improvisation through an algorithmic lens ...
Jazz improvisation mixes melodic and rhythmic constraints with an imperative for novelty and surprise. Precisely formalizing such constraints and developing algorithms for generating music within them is a lovely confluence of music, algorithms, and formal logic. I'd like to develop a CS + Music course around this connection. - What is "creativity"?
Ongoing Projects
- Emergence of abstraction: exploring a computational model in which phenomena associated with "abstraction" and symbolic languages can emerge as outcomes of simple computations
- DSL synthesis: theory and algorithms for automatically synthesizing DSLs that express relevant domain concepts succinctly and irrelevant ones less succinctly or not at all
Publications
*equal contribution