Research map / Architectures and efficiency / Neural architecture search
Progressive Neural Architecture Generation
Caiyang Yu, Chen Huang, Yun Liu, Chenwei Tang, Wei Ju, Jiancheng Lv
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 |
|---|---|---|---|
| replace diffusion-style iterative network inference with quantization-based multi-scale sub-architecture generation (MSQ) | PNAG vs DiffusionNAG · NB201 CIFAR-10 generation | time per architecture: 8 s -> 0.006 s (1300x) | Improved |
| add step-wise consistency constraint (SCC) on more generation steps (ablation) | PNAG · CIFAR10 on NB201 and MBV3 | validity and accuracy | Improved |
| use PNAG instead of DiffusionNAG for generation | NB201 Aircraft · NB201 Aircraft | accuracy (%): 58.83 -> 66.99 | Improved |
| use PNAG instead of DiffusionNAG | MBV3 CIFAR100 · MBV3 CIFAR100 | accuracy (%): 86.07 -> 87.20 | Improved |
| use PNAG instead of DiffusionNAG on CIFAR-10 and CIFAR-100 in NB201 | NB201 · NB201 CIFAR-10/CIFAR-100 | accuracy (%): 94.37 -> 94.37; 73.51 -> 73.51 | No clear effect |
| use PNAG instead of DiffusionNAG for DARTS-space CIFAR-100 | DARTS space · DARTS space CIFAR-100 | test accuracy (%): PNAG 83.54 vs OStr-DARTS 84.22 | Mixed |
| add step-wise validity supervision (PNAG) vs final-step-only constraint (DiffusionNAG) | PNAG · NB201 validity | validity (%): 98.97 -> 100.00 | Improved |
| swap surrogate model to match target objective (fClean, fAPGD, fBlur) | PNAG conditional generation · NB201 CIFAR10 Clean/APGD/Blur | max accuracy (%): APGD 34.52; Blur 48.37; Clean 92.17 | Improved |
| apply PNAG to Transformer search space (AutoFormer) Tiny | AutoFormer space · ImageNet Tiny | top-1 accuracy (%): 74.7 -> 76.6 | Improved |
| apply PNAG to AutoFormer Base | AutoFormer space · ImageNet Base | top-1 accuracy (%): 82.4 -> 82.4 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- AZ-NAS — AZ-NAS: Assembling Zero-Cost Proxies for Network Architecture Search (CVPR 2024) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Prior Knowledge Guided Neural Architecture Generation (ICML 2025 · similarity 0.87)
- Neural Architecture Generator Optimization (NeurIPS 2020 · similarity 0.87)
- ProxyBO: Accelerating Neural Architecture Search via Bayesian Optimization with Zero-Cost Proxies (AAAI 2023 · similarity 0.86)
- DiffusionNAG: Predictor-guided Neural Architecture Generation with Diffusion Models (ICLR 2024 · similarity 0.85)
- Towards Neural Architecture Search through Hierarchical Generative Modeling (ICML 2024 · similarity 0.85)
- EG-NAS: Neural Architecture Search with Fast Evolutionary Exploration (AAAI 2024 · similarity 0.85)
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.
