AI RESEARCH
SharpNet: Enhancing MLPs to Represent Functions with Controlled Non-differentiability
arXiv CS.CV
•
ArXi:2601.19683v2 Announce Type: replace Multi-layer perceptrons (MLPs) are a standard tool for learning and function approximation, but they inherently produce globally smooth outputs. Consequently, they struggle to represent functions that are continuous yet intentionally non-differentiable (i.e., functions with prescribed $C^0$ sharp features) without ad hoc post-processing.