SOURCE-LINKED INTELLIGENCE

Why and When Neural Networks Improve Local Approximation in Optimization

arXiv · AI, language, vision and robotics · article · Aug 25, 2026 · UTC

Published experience with neural surrogates in derivative-free optimisation is contradictory: the same family of models that cuts the evaluation count of one solver leaves another unchanged, or makes it worse. We show that the contradiction dissolves once three factors are stated, and that these, rather than the fit accuracy a training curve reports, are what delimit when a learned local model pays. Role: a surrogate that proposes candidates the true objective must still approve helps, while one that replaces a gradient the solver depends on hurts. Radius: a model fitted to an optimisation pat

Read original source ↗ Open in workspace

recordType

paper

region

Global