Zero-Cost Operation Scoring in Differentiable Architecture Search
DOI:
https://doi.org/10.1609/aaai.v37i9.26243Keywords:
ML: Auto ML and Hyperparameter Tuning, ML: Deep Neural ArchitecturesAbstract
We formalize and analyze a fundamental component of dif- ferentiable neural architecture search (NAS): local “opera- tion scoring” at each operation choice. We view existing operation scoring functions as inexact proxies for accuracy, and we find that they perform poorly when analyzed empir- ically on NAS benchmarks. From this perspective, we intro- duce a novel perturbation-based zero-cost operation scor- ing (Zero-Cost-PT) approach, which utilizes zero-cost prox- ies that were recently studied in multi-trial NAS but de- grade significantly on larger search spaces, typical for dif- ferentiable NAS. We conduct a thorough empirical evalu- ation on a number of NAS benchmarks and large search spaces, from NAS-Bench-201, NAS-Bench-1Shot1, NAS- Bench-Macro, to DARTS-like and MobileNet-like spaces, showing significant improvements in both search time and accuracy. On the ImageNet classification task on the DARTS search space, our approach improved accuracy compared to the best current training-free methods (TE-NAS) while be- ing over 10× faster (total searching time 25 minutes on a single GPU), and observed significantly better transferabil- ity on architectures searched on the CIFAR-10 dataset with an accuracy increase of 1.8 pp. Our code is available at: https://github.com/zerocostptnas/zerocost operation score.Downloads
Published
2023-06-26
How to Cite
Xiang, L., Dudziak, L., Abdelfattah, M. S., Chau, T., Lane, N. D., & Wen, H. (2023). Zero-Cost Operation Scoring in Differentiable Architecture Search. Proceedings of the AAAI Conference on Artificial Intelligence, 37(9), 10453-10463. https://doi.org/10.1609/aaai.v37i9.26243
Issue
Section
AAAI Technical Track on Machine Learning IV