Papers › IS-DARTS: Stabilizing DARTS through Precise Measurement on Candidate Importance
IS-DARTS: Stabilizing DARTS through Precise Measurement on Candidate Importance
Hongyi He, Longjun Liu, Haonan Zhang, Nanning Zheng
Among existing Neural Architecture Search methods, DARTS is known for its efficiency and simplicity. This approach applies continuous relaxation of network representation to construct a weight-sharing supernet and enables the identification of excellent subnets in just a few GPU days. However, performance collapse in DARTS results in deteriorating architectures filled with parameter-free operations and remains a great challenge to the robustness. To resolve this problem, we reveal that the fundamental reason is the biased estimation of the candidate importance in the search space through theoretical and experimental analysis, and more precisely select operations via information-based measurements. Furthermore, we demonstrate that the excessive concern over the supernet and inefficient utilization of data in bi-level optimization also account for suboptimal results. We adopt a more realistic objective focusing on the performance of subnets and simplify it with the help of the information-based measurements. Finally, we explain theoretically why progressively shrinking the width of the supernet is necessary and reduce the approximation error of optimal weights in DARTS. Our proposed method, named IS-DARTS, comprehensively improves DARTS and resolves the aforementioned problems. Extensive experiments on NAS-Bench-201 and DARTS-based search space demonstrate the effectiveness of IS-DARTS.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
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
1 archive task tag without a task page not shown.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Neural Architecture Search | NAS-Bench-201, CIFAR-10 | IS-DARTS | Accuracy (Test) | 94.36 | #7 of 37 | Archive leaderboard | report |
| Neural Architecture Search | NAS-Bench-201, CIFAR-10 | IS-DARTS | Accuracy (Val) | 91.55 | #7 of 37 | Archive leaderboard | report |
| Neural Architecture Search | NAS-Bench-201, CIFAR-100 | IS-DARTS | Accuracy (Test) | 73.51 | #7 of 40 | Archive leaderboard | report |
| Neural Architecture Search | NAS-Bench-201, CIFAR-100 | IS-DARTS | Accuracy (Val) | 73.49 | #7 of 40 | Archive leaderboard | report |
| Neural Architecture Search | NAS-Bench-201, ImageNet-16-120 | IS-DARTS | Accuracy (Test) | 46.34 | #14 of 49 | Archive leaderboard | report |
| Neural Architecture Search | NAS-Bench-201, ImageNet-16-120 | IS-DARTS | Accuracy (Val) | 46.37 | #14 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.
Methods
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