Research map / Architectures and efficiency / Neural architecture search
BossNAS: Exploring Hybrid CNN-Transformers With Block-Wisely Self-Supervised Neural Architecture Search
Changlin Li, Tao Tang, Guangrun Wang, Jiefeng Peng, Bing Wang, Xiaodan Liang, Xiaojun Chang
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 naive unsupervised bootstrapping with ensemble bootstrapping for supernet training (ablation) | BossNAS · MBConv search space rating | Kendall tau: 0.12 -> 0.65 | Improved |
| use unsupervised evaluation metric instead of supervised linear evaluation after ensemble bootstrapping (ablation) | BossNAS · MBConv search space rating | Kendall tau: 0.55 -> 0.65 | Improved |
| use unsupervised ensemble bootstrapping instead of supervised distillation (ablation) | BossNAS · MBConv search space rating | Kendall tau: 0.62 -> 0.65 | Improved |
| use unsupervised ensemble bootstrapping instead of supervised classification training (ablation) | BossNAS · MBConv search space rating | Kendall tau: 0.46 -> 0.65 | Improved |
| use a teacher unlike the candidates (MobileNetV1 instead of EfficientNet-style teacher) in DNA | DNA · MBConv search space rating | Kendall tau: 0.62 -> 0.23 | Hurt |
| search with BossNAS instead of supervised DNA-T on HyTra | HyTra search · ImageNet val (HyTra search space) | top-1 accuracy (%): 80.3 -> 80.8 | Improved |
| search with BossNAS instead of random selection in HyTra | HyTra search · ImageNet val (HyTra search space) | top-1 accuracy (%): 76.7 -> 80.8 | Improved |
| search with BossNAS instead of UnNAS in HyTra | HyTra search · ImageNet val (HyTra search space) | top-1 accuracy (%): 79.8 -> 80.8 | Improved |
| replace relative position encoding in BoTBlock with depthwise conv implicit position encoding (BoT50 to BoT50-T) | BoT50 · ImageNet val (HyTra search space) | top-1 accuracy (%): 78.3 -> 79.5 | Improved |
| add SE and SiLU to BossNet-T0 (ablation) | BossNet-T0 · ImageNet val (HyTra search space) | top-1 accuracy (%): 80.5 -> 80.8 | Improved |
| test BossNet-T0 on larger input (288) without finetuning | BossNet-T0 · ImageNet val (HyTra search space) | top-1 accuracy (%): 80.8 -> 81.6 | Improved |
| remove downsampling in last stage of BossNet-T0 (BossNet-T1) | BossNet-T0 · ImageNet val (HyTra search space) | top-1 accuracy (%): 80.8 -> 82.2 | Improved |
Closest papers by meaning
- BossNAS Family: Block-Wisely Self-Supervised Neural Architecture Search (TPAMI 2025 · similarity 0.96)
- Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspective (ICLR 2021 · similarity 0.90)
- TransNAS-Bench-101: Improving Transferability and Generalizability of Cross-Task Neural Architecture Search (CVPR 2021 · similarity 0.90)
- NAS-Bench-360: Benchmarking Neural Architecture Search on Diverse Tasks (NeurIPS 2022 · similarity 0.90)
- BigNAS: Scaling Up Neural Architecture Search with Big Single-Stage Models (ECCV 2020 · similarity 0.90)
- einspace: Searching for Neural Architectures from Fundamental Operations (NeurIPS 2024 · similarity 0.88)
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.
