Saved in:
Bibliographic Details
Main Authors: Carrasco, Pablo, Muñoz, Gonzalo
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.23362
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917370459586560
author Carrasco, Pablo
Muñoz, Gonzalo
author_facet Carrasco, Pablo
Muñoz, Gonzalo
contents The non-convex nature of trained neural networks has created significant obstacles in their incorporation into optimization models. In this context, Anderson et al. (2020) provided a framework to obtain the convex hull of the graph of a piecewise linear convex activation function composed with an affine function; this effectively convexifies activations such as the ReLU together with the affine transformation that precedes it. In this article, we contribute to this line of work by developing a recursive formula that yields a tight convexification for the composition of an activation with an affine function for a wide scope of activation functions, namely, convex or ``S-shaped". Our approach can be used to efficiently compute separating hyperplanes or determine that none exists in various settings, including non-polyhedral cases.
format Preprint
id arxiv_https___arxiv_org_abs_2410_23362
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Tightening convex relaxations of trained neural networks: a unified approach for convex and S-shaped activations
Carrasco, Pablo
Muñoz, Gonzalo
Optimization and Control
Machine Learning
The non-convex nature of trained neural networks has created significant obstacles in their incorporation into optimization models. In this context, Anderson et al. (2020) provided a framework to obtain the convex hull of the graph of a piecewise linear convex activation function composed with an affine function; this effectively convexifies activations such as the ReLU together with the affine transformation that precedes it. In this article, we contribute to this line of work by developing a recursive formula that yields a tight convexification for the composition of an activation with an affine function for a wide scope of activation functions, namely, convex or ``S-shaped". Our approach can be used to efficiently compute separating hyperplanes or determine that none exists in various settings, including non-polyhedral cases.
title Tightening convex relaxations of trained neural networks: a unified approach for convex and S-shaped activations
topic Optimization and Control
Machine Learning
url https://arxiv.org/abs/2410.23362