AI RESEARCH
Every Feedforward Neural Network Definable in an o-Minimal Structure Has Finite Sample Complexity
arXiv CS.LG
•
ArXi:2605.07097v1 Announce Type: cross We show that, in a precise sense, a broad class of feedforward neural networks learn (have finite sample complexity) in the PAC model: every fixed finite feedforward architecture whose layers are definable in an o-minimal structure has finite sample complexity in the agnostic PAC setting, even with unbounded parameters.