Research map / Architectures and efficiency / Neural architecture search
Distribution Consistent Neural Architecture Search
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use distribution-consistent local weight sharing | one-shot NAS on ImageNet under a 200-350M FLOPs constraint · ImageNet | Top-1 accuracy: 77.6 -> 78.1 | Helped |
| use distribution-consistent local weight sharing | one-shot NAS on ImageNet under a 350-500M FLOPs constraint · ImageNet | Top-1 accuracy: 79.2 -> 79.5 | Helped |
| use distribution-consistent local weight sharing | NAS-Bench-201 search · CIFAR-10 test | accuracy: 93.80±0.40 -> 94.29±0.07 | Helped |
| use distribution-consistent local weight sharing | NAS-Bench-201 search · CIFAR-100 test | accuracy: 71.53±1.51 -> 73.02±0.16 | Helped |
| use distribution-consistent local weight sharing | NAS-Bench-201 search · ImageNet16-120 test | accuracy: 45.12±0.82 -> 46.41±0.14 | Helped |
| replace learned local weight sharing with global weight sharing (ablation) | the proposed NAS method · ImageNet | Top-1 accuracy: 78.1 -> 77.2 | Hurt |
| assign local weight-sharing states randomly rather than learning them (ablation) | the proposed NAS method · ImageNet | Top-1 accuracy: 78.1 -> 77.4 | Hurt |
| add a second candidate weight state per operator instead of using one (ablation) | the proposed NAS method · ImageNet | Top-1 accuracy: 77.2 -> 78.1 | Helped |
| increase candidate weight states per operator from 2 to 3 (ablation) | the proposed NAS method · ImageNet | Top-1 accuracy: 78.1 -> 77.9 | Hurt |
| increase the candidate architecture pool from 5,000 to 10,000 (ablation) | the proposed NAS method · ImageNet | Top-1 accuracy: 77.7 -> 78.1 | Helped |
| increase the candidate architecture pool from 10,000 to 20,000 (ablation) | the proposed NAS method · ImageNet | Top-1 accuracy: 78.1 -> 78.0 | No clear effect |
Closest papers by meaning
- Efficient Few-Shot Neural Architecture Search by Counting the Number of Nonlinear Functions (AAAI 2025 · similarity 0.92)
- Fitting the Search Space of Weight-sharing NAS with Graph Convolutional Networks (AAAI 2021 · similarity 0.92)
- Landmark Regularization: Ranking Guided Super-Net Training in Neural Architecture Search (CVPR 2021 · similarity 0.92)
- Boosting Order-Preserving and Transferability for Neural Architecture Search: a Joint Architecture Refined Search and Fine-tuning Approach (CVPR 2024 · similarity 0.91)
- Subnet-Aware Dynamic Supernet Training for Neural Architecture Search (CVPR 2025 · similarity 0.90)
- HEP-NAS: Towards Efficient Few-shot Neural Architecture Search via Hierarchical Edge Partitioning (AAAI 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.