LESS：分钟级轻量进化超网搜索

View PDF HTML (experimental)

Abstract: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 \(93.189\pm0.467\%\) 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 \(69.615\pm1.139\%\) and \(43.720\pm1.697\%\) accuracy. Applied without tuning to the larger DARTS space, LESS achieves \(96.95\pm0.14\%\) on CIFAR-10 and \(82.43\pm0.80\%\) 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.

Comments:	33 pages, 4 figures. Code: this https URL
Subjects:	Neural and Evolutionary Computing (cs.NE); Machine Learning (cs.LG)
Cite as:	arXiv:2610.01468 [cs.NE]
	(or arXiv:2610.01468v1 [cs.NE] for this version)
	https://doi.org/10.48550/arXiv.2610.01468 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Aviral Gandhi [view email]
[v1] Thu, 1 Oct 2026 11:04:47 UTC (158 KB)