Research map / Architectures and efficiency / Neural architecture search
Neural Architecture Optimization
Read the paperCheck 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 |
|---|---|---|---|
| optimize architecture embeddings by gradient ascent and decode them into architectures | NAO · CIFAR-10 test | error rate (%): 3.34→ 3.18 | Helped |
| optimize architecture embeddings by gradient ascent and decode them into architectures | NAO · PTB test | perplexity: 56.1 to 56.0 | Helped |
| add symmetric architecture examples during encoder and predictor training (ablation) | NAO performance predictor · CIFAR-10 predictor evaluation | pairwise accuracy: about 2% | Helped |
| add Cutout (ablation) | NAONet · CIFAR-10 test | error rate (%): 3.18 -> 2.48 | Helped |
| use weight sharing (ablation) | NAONet · CIFAR-10 test | error rate (%): 3.18 -> 3.53 | Mixed |
| use NAO search instead of random search with weight sharing | weight-sharing architecture search · CIFAR-10 test | error rate (%): 3.92 -> 3.53 | Helped |
| use weight sharing (ablation) | NAONet + weight tying + weight penalty · PTB test | perplexity: 56.0 -> 56.6 | Hurt |
| use NAO search instead of random search with weight sharing | weight-sharing architecture search · PTB test | perplexity: 58.81 -> 56.6 | Helped |
| transfer the CIFAR-10-discovered architecture to CIFAR-100 | NAONet · CIFAR-100 test | error rate (%): 15.20 -> 14.75 | Helped |
| transfer the PTB-discovered architecture to WikiText-2 | NAONet · WikiText-2 test | perplexity: 66.9 vs. 67.0 | Mixed |
Closest papers by meaning
- Hierarchical Representations for Efficient Architecture Search (ICLR 2018 · similarity 0.89)
- A Semi-Supervised Assessor of Neural Architectures (CVPR 2020 · similarity 0.89)
- Building Optimal Neural Architectures using Interpretable Knowledge (CVPR 2024 · similarity 0.89)
- Encodings for Prediction-based Neural Architecture Search (ICML 2024 · similarity 0.88)
- Insights from the Use of Previously Unseen Neural Architecture Search Datasets (CVPR 2024 · similarity 0.87)
- EG-NAS: Neural Architecture Search with Fast Evolutionary Exploration (AAAI 2024 · similarity 0.87)
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.