Structure-Preserving Neural Approximation of Convex, Semiconvex, and Quasiconvex Functions
Organizations: Sejong University, 209 Neungdong-ro, Gwangjin-gu, Seoul, 05006, Seoul, South Korea · Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul, 03722, Seoul, South Korea
Abstract
Input convex neural networks provide a natural way to enforce convexity globally through architectural constraints. We study how this structure-preserving approximation framework extends to semiconvex, quasiconvex, and convex transformable functions. For convex functions on compact convex domains, we establish the sharp minimax rate under max-affine approximation with affine pieces, and show that every max-affine function admits an exact bounded-width ICNN representation with complexity linear in . The same approximation rate extends to semiconvex functions through quadratic lifting and to convex transformable functions through monotone composition. For general Lipschitz quasiconvex functions, we develop a quantitative finite-probe approximation theory based on monotone responses to affine projections and obtain a structure-preserving IqCNN with an approximation guarantee. We further show that these approximation results extend to ReLU-transferable smooth activations, including Softplus and ELU. Numerical experiments on PDE-based approximation problems compare the proposed architectures with parameter-matched MLPs and validation-tuned shape penalties, demonstrating global structural validity without penalty tuning while maintaining competitive approximation accuracy. Overall, the results identify recoverable convex latent structure as the key mechanism enabling gradient-level structure-preserving approximation beyond convexity.