← 研踪 PaperTrailarXiv:2611.24325 · v1
Structure-Preserving Neural Approximation of Convex, Semiconvex, and Quasiconvex Functions
首次提交 2026-11-18 20:29
版本更新 2026-11-18 20:29
雷达观测 2026-11-19 15:04
摘要
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.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-18 20:29 · 最新观测版本
Structure-Preserving Neural Approximation of Convex, Semiconvex, and Quasiconvex Functions
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.
查看此版本解读 →查看此版本原文 ↗