Understanding and Accelerating Neural Architecture Search with Training-Free and Theory-Grounded Metrics
- Publisher:
- IEEE COMPUTER SOC
- Publication Type:
- Journal Article
- Citation:
- IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46, (2), pp. 749-763
- Issue Date:
- 2024-02-01
Open Access
Copyright Clearance Process
- Recently Added
- In Progress
- Open Access
This item is open access.
This work targets designing a principled and unified training-free framework for Neural Architecture Search (NAS), with high performance, low cost, and in-depth interpretation. NAS has been explosively studied to automate the discovery of top-performer neural networks, but suffers from heavy resource consumption and often incurs search bias due to truncated training or approximations. Recent NAS works Mellor et al. 2021, Chen et al. 2021, Abdelfattah et al. 2021 start to explore indicators that can predict a network's performance without training. However, they either leveraged limited properties of deep networks, or the benefits of their training-free indicators were not applied to more extensive search methods. By rigorous correlation analysis, we present a unified framework to understand and accelerate NAS, by disentangling 'TEG' characteristics of searched networks - Trainability, Expressivity, Generalization - all assessed in a training-free manner. The TEG indicators could be scaled up and integrated with various NAS search methods, including both supernet and single-path NAS approaches. Extensive studies validate the effective and efficient guidance from our TEG-NAS framework, leading to both improved search accuracy and over 56% reduction in search time cost. Moreover, we visualize search trajectories on three landscapes of 'TEG' characteristics, observing that a good local minimum is easier to find on NAS-Bench-201 given its simple topology, whereas balancing 'TEG' characteristics is much harder on the DARTS space due to its complex landscape geometry.
Please use this identifier to cite or link to this item: