A Novel Framework for Neural Architecture Search in the Hill Climbing Domain

TL;DR

This paper introduces a new neural architecture search framework using hill climbing and morphism operators, which reduces training time and achieves competitive results on CIFAR-10.

Contribution

A novel hill-climbing based neural architecture search method employing morphism operators and a gradient update scheme that accelerates training and broadens search space.

Findings

Achieved 4.96% error rate on CIFAR-10

Reduced training time to 19.4 hours on a single GPU

Broader search space yields competitive architectures

Abstract

Neural networks have now long been used for solving complex problems of image domain, yet designing the same needs manual expertise. Furthermore, techniques for automatically generating a suitable deep learning architecture for a given dataset have frequently made use of reinforcement learning and evolutionary methods which take extensive computational resources and time. We propose a new framework for neural architecture search based on a hill-climbing procedure using morphism operators that makes use of a novel gradient update scheme. The update is based on the aging of neural network layers and results in the reduction in the overall training time. This technique can search in a broader search space which subsequently yields competitive results. We achieve a 4.96% error rate on the CIFAR-10 dataset in 19.4 hours of a single GPU training.