Papers › iDARTS: Improving DARTS by Node Normalization and Decorrelation Discretization

iDARTS: Improving DARTS by Node Normalization and Decorrelation Discretization

25 Aug 2021arXiv:2108.11014archive 2025-07-28

Huiqun Wang, Ruijie Yang, Di Huang, Yunhong Wang

Differentiable ARchiTecture Search (DARTS) uses a continuous relaxation of network representation and dramatically accelerates Neural Architecture Search (NAS) by almost thousands of times in GPU-day. However, the searching process of DARTS is unstable, which suffers severe degradation when training epochs become large, thus limiting its application. In this paper, we claim that this degradation issue is caused by the imbalanced norms between different nodes and the highly correlated outputs from various operations. We then propose an improved version of DARTS, namely iDARTS, to deal with the two problems. In the training phase, it introduces node normalization to maintain the norm balance. In the discretization phase, the continuous architecture is approximated based on the similarity between the outputs of the node and the decorrelated operations rather than the values of the architecture parameters. Extensive evaluation is conducted on CIFAR-10 and ImageNet, and the error rates of 2.25\% and 24.7\% are reported within 0.2 and 1.9 GPU-day for architecture search respectively, which shows its effectiveness. Additional analysis also reveals that iDARTS has the advantage in robustness and generalization over other DARTS-based counterparts.

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

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

Tasks

Neural Architecture Search

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Neural Architecture Search CIFAR-10 iDARTS +ME Parameters 3.6M #9 of 41 Archive leaderboard report
Neural Architecture Search CIFAR-10 iDARTS +ME Search Time (GPU days) 0.4 #9 of 41 Archive leaderboard report
Neural Architecture Search CIFAR-10 iDARTS +ME Top-1 Error Rate 2.25% #9 of 41 Archive leaderboard report
Neural Architecture Search ImageNet iDARTS (ImageNet) MACs 568M #112 of 135 Archive leaderboard report
Neural Architecture Search ImageNet iDARTS (ImageNet) Params 5.1M #112 of 135 Archive leaderboard report
Neural Architecture Search ImageNet iDARTS (ImageNet) Top-1 Error Rate 24.7 #112 of 135 Archive leaderboard report
Neural Architecture Search ImageNet iDARTS (CIFAR-10) MACs 578M #117 of 135 Archive leaderboard report
Neural Architecture Search ImageNet iDARTS (CIFAR-10) Params 5.1M #117 of 135 Archive leaderboard report
Neural Architecture Search ImageNet iDARTS (CIFAR-10) Top-1 Error Rate 25.2 #117 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

DARTS

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