DAAS: Differentiable Architecture and Augmentation Policy Search release_qvet6o3mqjgr7dtgfjakero3iu

by Xiaoxing Wang, Xiangxiang Chu, Junchi Yan, Xiaokang Yang

Released as a article .

2022  

Abstract

Neural architecture search (NAS) has been an active direction of automatic machine learning (Auto-ML), aiming to explore efficient network structures. The searched architecture is evaluated by training on datasets with fixed data augmentation policies. However, recent works on auto-augmentation show that the suited augmentation policies can vary over different structures. Therefore, this work considers the possible coupling between neural architectures and data augmentation and proposes an effective algorithm jointly searching for them. Specifically, 1) for the NAS task, we adopt a single-path based differentiable method with Gumbel-softmax reparameterization strategy due to its memory efficiency; 2) for the auto-augmentation task, we introduce a novel search method based on policy gradient algorithm, which can significantly reduce the computation complexity. Our approach achieves 97.91% accuracy on CIFAR-10 and 76.6% Top-1 accuracy on ImageNet dataset, showing the outstanding performance of our search algorithm.
In text/plain format

Archived Files and Locations

application/pdf  402.9 kB
file_engtp7q5vjdejfmaeaaitmgn7m
arxiv.org (repository)
web.archive.org (webarchive)
Read Archived PDF
Preserved and Accessible
Type  article
Stage   submitted
Date   2022-01-28
Version   v2
Language   en ?
arXiv  2109.15273v2
Work Entity
access all versions, variants, and formats of this works (eg, pre-prints)
Catalog Record
Revision: 9ab4dbab-9088-46b1-9e3a-7b2d6aecd3fa
API URL: JSON