Papers › DARTS-PRIME: Regularization and Scheduling Improve Constrained Optimization in...

DARTS-PRIME: Regularization and Scheduling Improve Constrained Optimization in Differentiable NAS

22 Jun 2021arXiv:2106.11655archive 2025-07-28

Kaitlin Maile, Erwan Lecarpentier, Hervé Luga, Dennis G. Wilson

Differentiable Architecture Search (DARTS) is a recent neural architecture search (NAS) method based on a differentiable relaxation. Due to its success, numerous variants analyzing and improving parts of the DARTS framework have recently been proposed. By considering the problem as a constrained bilevel optimization, we present and analyze DARTS-PRIME, a variant including improvements to architectural weight update scheduling and regularization towards discretization. We propose a dynamic schedule based on per-minibatch network information to make architecture updates more informed, as well as proximity regularization to promote well-separated discretization. Our results in multiple domains show that DARTS-PRIME improves both performance and reliability, comparable to state-of-the-art in differentiable NAS.

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

Bilevel OptimizationNeural Architecture SearchScheduling

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Neural Architecture Search CIFAR-10 DARTS-PRIME Parameters 3.7M #29 of 41 Archive leaderboard report
Neural Architecture Search CIFAR-10 DARTS-PRIME Search Time (GPU days) 0.5 #29 of 41 Archive leaderboard report
Neural Architecture Search CIFAR-10 DARTS-PRIME Top-1 Error Rate 2.62% #29 of 41 Archive leaderboard report
Neural Architecture Search CIFAR-100 DARTS-PRIME PARAMS 3.16M #10 of 13 Archive leaderboard report
Neural Architecture Search CIFAR-100 DARTS-PRIME Percentage Error 17.44 #10 of 13 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

DARTSProximity Regularization

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