Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Noisy Differentiable Architecture Search

About

Simplicity is the ultimate sophistication. Differentiable Architecture Search (DARTS) has now become one of the mainstream paradigms of neural architecture search. However, it largely suffers from the well-known performance collapse issue due to the aggregation of skip connections. It is thought to have overly benefited from the residual structure which accelerates the information flow. To weaken this impact, we propose to inject unbiased random noise to impede the flow. We name this novel approach NoisyDARTS. In effect, a network optimizer should perceive this difficulty at each training step and refrain from overshooting, especially on skip connections. In the long run, since we add no bias to the gradient in terms of expectation, it is still likely to converge to the right solution area. We also prove that the injected noise plays a role in smoothing the loss landscape, which makes the optimization easier. Our method features extreme simplicity and acts as a new strong baseline. We perform extensive experiments across various search spaces, datasets, and tasks, where we robustly achieve state-of-the-art results. Our code is available at https://github.com/xiaomi-automl/NoisyDARTS.

Xiangxiang Chu, Bo Zhang• 2020

Related benchmarks

TaskDatasetResultRank
Image ClassificationCIFAR-10 (test)
Accuracy97.63
3381
Object DetectionCOCO 2017 (val)
AP33.1
2454
Image ClassificationImageNet (val)
Top-1 Acc77.9
1206
Image ClassificationCIFAR-10 (test)
Accuracy97.53
906
Image ClassificationSVHN (test)
Accuracy97.67
362
Image ClassificationCIFAR-100 (test)
Top-1 Acc79.93
275
Image ClassificationImageNet (test)--
235
Image ClassificationCIFAR-10 NAS-Bench-201 (test)
Accuracy93.49
173
Image ClassificationCIFAR-100 NAS-Bench-201 (test)
Accuracy71.55
169
Image ClassificationImageNet-16-120 NAS-Bench-201 (test)
Accuracy42.34
139
Showing 10 of 15 rows

Other info

Code

Follow for update