Research map / Architectures and efficiency / Neural architecture search
Progressive DARTS: Bridging the Optimization Gap for NAS in the Wild
Xin Chen, Lingxi Xie, Jun Hui Wu, Qi Chuan Tian
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add operation-level Dropout on skip-connect (M=2) (ablation) | P-DARTS · CIFAR10 | test error (%): 2.93 -> 2.69 | Improved |
| add operation-level Dropout on skip-connect (M=3) (ablation) | P-DARTS · CIFAR10 | test error (%): 3.28 -> 2.84 | Improved |
| add operation-level Dropout on skip-connect (M=4) (ablation) | P-DARTS · CIFAR10 | test error (%): 3.51 -> 2.97 | Improved |
| approximate search space progressively instead of random sampling (ablation) | P-DARTS final stage · CIFAR10 | test error (%): 3.43 -> 2.58 | Improved |
| shrink operation set 8 to 5 to 3 at fixed depth of 8 cells (ablation) | DARTS first order · CIFAR10 | test error (%): 3.00 -> 2.70 | Improved |
| add architecture refinement limiting skip-connects to 2 (ablation) | P-DARTS · CIFAR10 | test error (%): 2.79 -> 2.65 | Improved |
| apply search space regularization to DARTS (ablation) | DARTS first order · CIFAR10 | test error (%): 3.00 -> 2.78 | Improved |
| increase number of preserved skip-connects from 0 to 4 (ablation) | P-DARTS · CIFAR10 | test error (%): 2.78, 2.68, 2.69, 2.84, 2.97 for M=0 to 4 | Mixed |
| progressively increase network width during search (ablation) | P-DARTS · CIFAR10 | test error (%) | No clear effect |
| add progressive depth search (P-DARTS) to DARTS | DARTS first order · CIFAR10 | test error (%): 3.00 -> 2.50 | Improved |
| search directly on CIFAR100 instead of transferring from CIFAR10 | P-DARTS · CIFAR100 | test error (%): 17.20 -> 15.92 | Improved |
| use large setting with 64 initial channels | P-DARTS (searched on CIFAR10) · CIFAR10 | test error (%): 2.50 -> 2.25 | Improved |
Closest papers by meaning
- Differentiable Architecture Search With Random Features (CVPR 2023 · similarity 0.92)
- iDARTS: Differentiable Architecture Search with Stochastic Implicit Gradients (ICML 2021 · similarity 0.91)
- $\beta $-DARTS++: Bi-Level Regularization for Proxy-Robust Differentiable Architecture Search (TPAMI 2025 · similarity 0.91)
- TransNAS-Bench-101: Improving Transferability and Generalizability of Cross-Task Neural Architecture Search (CVPR 2021 · similarity 0.91)
- Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspective (ICLR 2021 · similarity 0.91)
- EG-NAS: Neural Architecture Search with Fast Evolutionary Exploration (AAAI 2024 · 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.
