Steepest Descent Neural Architecture Optimization: Escaping Local Optimum with Signed Neural Splitting

TL;DR

This paper introduces a signed neural splitting method that enhances neural architecture optimization by escaping local optima, leading to improved performance on multiple benchmarks.

Contribution

It extends the splitting steepest descent framework by allowing signed splits, significantly improving optimization capability both theoretically and empirically.

Findings

Outperforms S2D and other methods on CIFAR-100, ImageNet, and ModelNet40.

Effectively escapes local optima in neural architecture search.

Achieves more accurate and energy-efficient neural networks.

Abstract

Developing efficient and principled neural architecture optimization methods is a critical challenge of modern deep learning. Recently, Liu et al.[19] proposed a splitting steepest descent (S2D) method that jointly optimizes the neural parameters and architectures based on progressively growing network structures by splitting neurons into multiple copies in a steepest descent fashion. However, S2D suffers from a local optimality issue when all the neurons become "splitting stable", a concept akin to local stability in parametric optimization. In this work, we develop a significant and surprising extension of the splitting descent framework that addresses the local optimality issue. The idea is to observe that the original S2D is unnecessarily restricted to splitting neurons into positive weighted copies. By simply allowing both positive and negative weights during splitting, we can eliminate the appearance of splitting stability in S2D and hence escape the local optima to obtain better performance. By incorporating signed splittings, we significantly extend the optimization power of splitting steepest descent both theoretically and empirically. We verify our method on various challenging benchmarks such as CIFAR-100, ImageNet and ModelNet40, on which we outperform S2D and other advanced methods on learning accurate and energy-efficient neural networks.