Papers › DrNAS: Dirichlet Neural Architecture Search

DrNAS: Dirichlet Neural Architecture Search

18 Jun 2020ICLR 2021 1arXiv:2006.10355archive 2025-07-28

Xiangning Chen, Ruochen Wang, Minhao Cheng, Xiaocheng Tang, Cho-Jui Hsieh

This paper proposes a novel differentiable architecture search method by formulating it into a distribution learning problem. We treat the continuously relaxed architecture mixing weight as random variables, modeled by Dirichlet distribution. With recently developed pathwise derivatives, the Dirichlet parameters can be easily optimized with gradient-based optimizer in an end-to-end manner. This formulation improves the generalization ability and induces stochasticity that naturally encourages exploration in the search space. Furthermore, to alleviate the large memory consumption of differentiable NAS, we propose a simple yet effective progressive learning scheme that enables searching directly on large-scale tasks, eliminating the gap between search and evaluation phases. Extensive experiments demonstrate the effectiveness of our method. Specifically, we obtain a test error of 2.46% for CIFAR-10, 23.7% for ImageNet under the mobile setting. On NAS-Bench-201, we also achieve state-of-the-art results on all three datasets and provide insights for the effective design of neural architecture search algorithms.

PaperPDFConference PDFCode

In Syntology View this paper on Syntology: its page in Syntology's graph, with its repositories and citations.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

xiangning-chen/DrNAS officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Neural Architecture Search

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Neural Architecture Search CIFAR-10 DrNAS Parameters 4.1M #17 of 41 Archive leaderboard report
Neural Architecture Search CIFAR-10 DrNAS Top-1 Error Rate 2.46% #17 of 41 Archive leaderboard report
Neural Architecture Search ImageNet DrNAS Params 5.7M #88 of 135 Archive leaderboard report
Neural Architecture Search ImageNet DrNAS Top-1 Error Rate 23.7 #88 of 135 Archive leaderboard report
Neural Architecture Search NAS-Bench-201, CIFAR-10 DrNAS Accuracy (Test) 94.36 #5 of 37 Archive leaderboard report
Neural Architecture Search NAS-Bench-201, CIFAR-10 DrNAS Accuracy (Val) 91.55 #5 of 37 Archive leaderboard report
Neural Architecture Search NAS-Bench-201, CIFAR-100 DrNAS Accuracy (Test) 73.51 #4 of 40 Archive leaderboard report
Neural Architecture Search NAS-Bench-201, CIFAR-100 DrNAS Accuracy (Val) 73.49 #4 of 40 Archive leaderboard report
Neural Architecture Search NAS-Bench-201, ImageNet-16-120 DrNAS Accuracy (Test) 46.34 #11 of 49 Archive leaderboard report
Neural Architecture Search NAS-Bench-201, ImageNet-16-120 DrNAS Accuracy (Val) 46.37 #11 of 49 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections