Papers › AtomNAS: Fine-Grained End-to-End Neural Architecture Search

AtomNAS: Fine-Grained End-to-End Neural Architecture Search

20 Dec 2019ICLR 2020 1arXiv:1912.09640archive 2025-07-28

Jieru Mei, Yingwei Li, Xiaochen Lian, Xiaojie Jin, Linjie Yang, Alan Yuille, Jianchao Yang

Search space design is very critical to neural architecture search (NAS) algorithms. We propose a fine-grained search space comprised of atomic blocks, a minimal search unit that is much smaller than the ones used in recent NAS algorithms. This search space allows a mix of operations by composing different types of atomic blocks, while the search space in previous methods only allows homogeneous operations. Based on this search space, we propose a resource-aware architecture search framework which automatically assigns the computational resources (e.g., output channel numbers) for each operation by jointly considering the performance and the computational cost. In addition, to accelerate the search process, we propose a dynamic network shrinkage technique which prunes the atomic blocks with negligible influence on outputs on the fly. Instead of a search-and-retrain two-stage paradigm, our method simultaneously searches and trains the target architecture. Our method achieves state-of-the-art performance under several FLOPs configurations on ImageNet with a small searching cost. We open our entire codebase at: https://github.com/meijieru/AtomNAS.

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

meijieru/AtomNAS officialmentioned in papermentioned on GitHubpytorchNOASSERTION 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 ImageNet AtomNAS-C+† Accuracy 77.6 #61 of 135 Archive leaderboard report
Neural Architecture Search ImageNet AtomNAS-C+† MACs 363M #61 of 135 Archive leaderboard report
Neural Architecture Search ImageNet AtomNAS-C+† Params 5.9M #61 of 135 Archive leaderboard report
Neural Architecture Search ImageNet AtomNAS-C+† Top-1 Error Rate 22.4 #61 of 135 Archive leaderboard report
Neural Architecture Search ImageNet AtomNAS-B+† Accuracy 77.2 #68 of 135 Archive leaderboard report
Neural Architecture Search ImageNet AtomNAS-B+† MACs 329M #68 of 135 Archive leaderboard report
Neural Architecture Search ImageNet AtomNAS-B+† Params 5.5M #68 of 135 Archive leaderboard report
Neural Architecture Search ImageNet AtomNAS-B+† Top-1 Error Rate 22.8 #68 of 135 Archive leaderboard report
Neural Architecture Search ImageNet AtomNAS-A+† Accuracy 76.3 #87 of 135 Archive leaderboard report
Neural Architecture Search ImageNet AtomNAS-A+† MACs 260M #87 of 135 Archive leaderboard report
Neural Architecture Search ImageNet AtomNAS-A+† Params 4.7M #87 of 135 Archive leaderboard report
Neural Architecture Search ImageNet AtomNAS-A+† Top-1 Error Rate 23.7 #87 of 135 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.

Methods

LSTMSigmoid ActivationSoftmaxTanh Activation

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