Pith. sign in

REVIEW 5 cited by

DARTS+: Improved Differentiable Architecture Search with Early Stopping

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1909.06035 v2 pith:E4CAYL7L submitted 2019-09-13 cs.CV cs.LG

Hanwen Liang , Shifeng Zhang , Jiacheng Sun , Xingqiu He , Weiran Huang , Kechen Zhuang , Zhenguo Li This is my paper
classification cs.CVcs.LG
keywords dartssearchearlystoppingarchitecturecollapsealgorithmcriterion
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

Recently, there has been a growing interest in automating the process of neural architecture design, and the Differentiable Architecture Search (DARTS) method makes the process available within a few GPU days. However, the performance of DARTS is often observed to collapse when the number of search epochs becomes large. Meanwhile, lots of "{\em skip-connect}s" are found in the selected architectures. In this paper, we claim that the cause of the collapse is that there exists overfitting in the optimization of DARTS. Therefore, we propose a simple and effective algorithm, named "DARTS+", to avoid the collapse and improve the original DARTS, by "early stopping" the search procedure when meeting a certain criterion. We also conduct comprehensive experiments on benchmark datasets and different search spaces and show the effectiveness of our DARTS+ algorithm, and DARTS+ achieves $2.32\%$ test error on CIFAR10, $14.87\%$ on CIFAR100, and $23.7\%$ on ImageNet. We further remark that the idea of "early stopping" is implicitly included in some existing DARTS variants by manually setting a small number of search epochs, while we give an {\em explicit} criterion for "early stopping".

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. DASViT: Differentiable Architecture Search for Vision Transformer

    cs.LG 2025-07 conditional novelty 6.0 of 10

    DASViT searches Vision Transformer encoder topologies with a gradient-based DARTS approach and reports architectures that outperform ViT-B/16 on CIFAR-10, CIFAR-100, and ImageNet-100 without pre-training.

  2. Harvesting AI Computation at the Edge via Generic Approximation

    cs.AR 2026-06 unverdicted novelty 5.0 of 10

    A framework converts traditional edge tasks to NN models via NAS and schedules them on idle AI chips to improve performance without affecting primary workloads.

  3. confopt: A Library for Implementation and Evaluation of Gradient-based One-Shot NAS Methods

    cs.LG 2025-07 conditional novelty 5.0 of 10

    A library and nine DARTS-derived benchmarks show that relative rankings of gradient-based one-shot NAS methods are unstable, making DARTS-only evaluation unreliable.

  4. Implantable Adaptive Cells: A Novel Enhancement for Pre-Trained U-Nets in Medical Image Segmentation

    cs.CV 2024-05 unverdicted novelty 5.0 of 10

    Introduces Implantable Adaptive Cells inserted into pre-trained U-Nets via Partially-Connected DARTS to achieve approximately 5 percentage point gains in segmentation accuracy on four medical MRI/CT datasets.

  5. Bilevel Optimization for Neural Architecture Search

    cs.LG 2026-06 unverdicted novelty 3.0 of 10

    Reviews NAS methods through bilevel optimization lens, categorizing them into sampling-based and theory-based, and proposes an auxiliary math programming framework for more principled architecture and weight updates.

Pith tools