mlsanityData to 7 Oct 2026

LESS: Lightweight Evolutionary Supernet Search in Minutes

arXiv 2610.01468 ↗First posted 1 Oct 2026cs.NE cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Low-cost NAS must both explore high-performing architectures and identify them reliably, yet reducing evaluation cost often weakens the fidelity of candidate comparisons. Training-free methods reduce evaluation cost by replacing learned task feedback with proxy signals measured at initialization. We introduce LESS (Lightweight Evolutionary Supernet Search), a data-driven method that combines a brief fair hard-path warm-up with discrete search under a single CMA-ES distribution. Each proposal is evaluated as its decoded hard genotype after six candidate-conditioned supernet updates. On NAS-Bench-201, LESS achieves (93.189±0.467%) CIFAR-10 test accuracy in 409.1 seconds, coming within 0.04 percentage points of FairNAS using approximately (1/24) of its source-reported search time. Matched controls show that calibration improves selected validation accuracy by (0.577) percentage points while changing best-visited accuracy by only (0.054) points, indicating that its primary effect is to reduce selection regret. The frozen configuration transfers without tuning to CIFAR-100 and ImageNet16-120 with (69.615±1.139%) and (43.720±1.697%) accuracy. Applied without tuning to the larger DARTS space, LESS achieves (96.95±0.14%) on CIFAR-10 and (82.43±0.80%) on CIFAR-100, with each search completing in approximately 43.5 minutes on a single GPU. Together, these results show that short, balanced, data-dependent updates enable competitive neural architecture search across datasets and search spaces within minutes.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 32 of its 34 references that are in the map. First posted on arXiv 1 Oct 2026.

Builds on

32 of 34 references.

References · 32

  1. 2018DARTS: Differentiable Architecture Searchcited by 1,782
  2. 2017Learning Transferable Architectures for Scalable Image Recognitioncited by 1,097
  3. 2018Regularized Evolution for Image Classifier Architecture Searchcited by 902
  4. 2018Efficient Neural Architecture Search via Parameter Sharingcited by 831
  5. 2019Once-for-All: Train One Network and Specialize it for Efficient Deploymentcited by 778
  6. 2018BOHB: Robust and Efficient Hyperparameter Optimization at Scalecited by 466
  7. 2020NAS-Bench-201: Extending the Scope of Reproducible Neural Architecture Searchcited by 412
  8. 2019Single Path One-Shot Neural Architecture Search with Uniform Samplingcited by 411
  9. 2018SNAS: Stochastic Neural Architecture Searchcited by 288
  10. 2019PC-DARTS: Partial Channel Connections for Memory-Efficient Architecture Searchcited by 265
  11. 2019Searching for A Robust Neural Architecture in Four GPU Hourscited by 258
  12. 2019Random Search and Reproducibility for Neural Architecture Searchcited by 253
20 more
  1. 2020Neural Architecture Search without Trainingcited by 241
  2. 2019Progressive Differentiable Architecture Search: Bridging the Depth Gap between Search and Evaluationcited by 230
  3. 2021Zero-Cost Proxies for Lightweight NAScited by 194
  4. 2019Understanding and Robustifying Differentiable Architecture Searchcited by 176
  5. 2021Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspectivecited by 168
  6. 2018Neural Architecture Optimizationcited by 164
  7. 2019FairNAS: Rethinking Evaluation Fairness of Weight Sharing Neural Architecture Searchcited by 129
  8. 2019Fair DARTS: Eliminating Unfair Advantages in Differentiable Architecture Searchcited by 129
  9. 2021Rethinking Architecture Selection in Differentiable NAScited by 92
  10. 2021How Powerful are Performance Predictors in Neural Architecture Search?cited by 88
  11. 2020EagleEye: Fast Sub-net Evaluation for Efficient Neural Network Pruningcited by 80
  12. 2020GreedyNAS: Towards Fast One-Shot NAS with Greedy Supernetcited by 69
  13. 2023ZiCo: Zero-shot NAS via Inverse Coefficient of Variation on Gradientscited by 57
  14. 2020Few-shot Neural Architecture Searchcited by 50
  15. 2022NAS-Bench-Suite-Zero: Accelerating Research on Zero Cost Proxiescited by 36
  16. 2022Generalizing Few-Shot NAS with Gradient Matchingcited by 25
  17. 2024AZ-NAS: Assembling Zero-Cost Proxies for Network Architecture Searchcited by 25
  18. 2020Speedy Performance Estimation for Neural Architecture Searchcited by 24
  19. 2024Boosting Order-Preserving and Transferability for Neural Architecture Search: a Joint Architecture Refined Search and Fine-tuning Approachcited by 4
  20. 2021Neural Architecture Search using Covariance Matrix Adaptation Evolution Strategycited by 2

Built on by

Not cited yet.