Showing math.CTShow all
2 papers · 1 filter
math.CT2026
Presenting Neural Networks via Coherent Functors
Matthew Pugh, Jo Grundy, Corina Cirstea +1
This paper develops a methodology for representing machine learning models as models of formal theories, grounded in the perspective that machine learning models are a form of data…
math.CT2025
Learning Is a Kan Extension
Matthew Pugh, Jo Grundy, Corina Cirstea +1
Previous work has demonstrated that efficient algorithms exist for computing Kan extensions and that some Kan extensions have interesting similarities to various machine learning a…