Research map / Architectures and efficiency / Neural architecture search
Homogeneous Architecture Augmentation for Neural Predictor
Yuqiao Liu, Yehui Tang, Yanan Sun
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 homogeneous architecture augmentation (ablation) | random forest predictor · NAS-Bench-101 (N_ot=424) | KTau: 0.5422 -> 0.6888 | Improved |
| add one-hot encoding (ablation) | random forest predictor · NAS-Bench-101 (N_ot=424) | KTau: 0.5422 -> 0.5909 | Improved |
| add one-hot encoding on top of augmentation (ablation) | random forest predictor · NAS-Bench-101 (N_ot=424) | KTau: 0.6888 -> 0.6991 | Improved |
| add homogeneous architecture augmentation (ablation) | random forest predictor · NAS-Bench-201 (N_ot=424) | KTau: 0.6919 -> 0.7824 | Improved |
| add homogeneous architecture augmentation (ablation) | random forest predictor · NAS-Bench-201 (N_ot=1K) | KTau: 0.7754 -> 0.8172 | Improved |
| add one-hot encoding (ablation) | random forest predictor · NAS-Bench-201 (N_ot=1K) | KTau: 0.7754 -> 0.7647 | Hurt |
| add one-hot encoding on top of augmentation (ablation) | random forest predictor · NAS-Bench-201 (N_ot=1K) | KTau: 0.8172 -> 0.8166 | No clear effect |
| add one-hot encoding on top of augmentation (ablation) | AdaBoost predictor · NAS-Bench-101 (N_ot=424) | KTau: 0.4185 -> 0.3309 | Hurt |
| add homogeneous augmentation (HA) to NPNAS (ablation) | NPNAS · NAS-Bench-101 (N_ot=424) | KTau: 0.6945 -> 0.7062 | Improved |
| use more augmented architectures for training (ablation) | HAAP · NAS-Bench-101 (N_ot=424) | KTau | Improved |
| use HAAP predictor instead of SSANA in EA search | EA search · NAS-Bench-101 search | accuracy (%): 94.01 -> 94.09 | Improved |
| use HAAP with half time budget instead of REA | EA search · NAS-Bench-201 CIFAR-10 test | accuracy (%): 93.92 -> 94.00 | Improved |
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.
- ReNAS — ReNAS: Relativistic Evaluation of Neural Architecture Search (CVPR 2021) · 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
- A Semi-Supervised Assessor of Neural Architectures (CVPR 2020 · similarity 0.88)
- DA-NAS: Data Adapted Pruning for Efficient Neural Architecture Search (ECCV 2020 · similarity 0.87)
- Encodings for Prediction-based Neural Architecture Search (ICML 2024 · similarity 0.87)
- Unifying and Boosting Gradient-Based Training-Free Neural Architecture Search (NeurIPS 2022 · similarity 0.87)
- HyperNAS: Enhancing Architecture Representation for NAS Predictor via Hypernetwork (CVPR 2026 · similarity 0.86)
- Bridge the Gap Between Architecture Spaces via A Cross-Domain Predictor (NeurIPS 2022 · similarity 0.86)
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.
