Research map / Label-efficient and robust learning / Recognition and long-tail
LA3: Efficient Label-Aware AutoAugment
Mingjun Zhao, Shan Lu, Zixuan Wang, Xiaoli Wang, Di Niu
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 label-aware policies (ablation) | LA3 · CIFAR-100 WRN-40-2 | top-1 accuracy (%): 80.08 -> 81.09 | Improved |
| add label-aware policies (ablation) | LA3 · CIFAR-10 WRN-28-10 | top-1 accuracy (%): 97.11 -> 97.80 | Improved |
| add mRMR policy construction instead of top-100 (ablation) | LA3 · CIFAR-100 WRN-40-2 | top-1 accuracy (%): 78.57 -> 81.09 | Improved |
| add mRMR policy construction instead of top-100 (ablation) | LA3 · CIFAR-10 WRN-28-10 | top-1 accuracy (%): 97.49 -> 97.80 | Improved |
| add mRMR policy construction instead of top-500 (ablation) | LA3 · CIFAR-100 WRN-28-10 | top-1 accuracy (%): 84.04 -> 84.54 | Improved |
| increase top-k from 100 to 500 (ablation) | LA3 top-k variant · CIFAR-100 WRN-40-2 | top-1 accuracy (%): 78.57 -> 79.85 | Improved |
| increase top-k from 100 to 500 (ablation) | LA3 top-k variant · CIFAR-10 WRN-28-10 | top-1 accuracy (%): 97.49 -> 97.26 | Hurt |
| add batch augment (x4) | LA3 · ImageNet ResNet-50 | top-1 accuracy (%): 78.71 -> 79.97 | Improved |
| use LA3 policy instead of FastAA | ResNet-50 · ImageNet search cost | search cost (GPU hours): 450 -> 29.3 GPU hours | Improved |
| use LA3 policy instead of DADA | ResNet-50 · ImageNet | top-1 accuracy (%): 77.5 -> 78.71 | Improved |
| use LA3 searched on WRN-40-2 instead of AutoAugment | Shake-Shake / PyramidNet · CIFAR-100 PyramidNet+ShakeDrop | top-1 accuracy (%): 89.3 -> 89.02 | Hurt |
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.
- DADA — DADA: Depth-Aware Domain Adaptation in Semantic Segmentation (ICCV 2019) · better on 7 of 7 shared results
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
- Adversarial Auto-Augment with Label Preservation: A Representation Learning Principle Guided Approach (NeurIPS 2022 · similarity 0.89)
- AutoAugment: Learning Augmentation Strategies From Data (CVPR 2019 · similarity 0.88)
- Faster AutoAugment: Learning Augmentation Strategies Using Backpropagation (ECCV 2020 · similarity 0.87)
- Sample-Aware RandAugment: Search-Free Automatic Data Augmentation for Effective Image Recognition (IJCV 2025 · similarity 0.87)
- Augmentation Strategies for Learning With Noisy Labels (CVPR 2021 · similarity 0.87)
- SAFLEX: Self-Adaptive Augmentation via Feature Label Extrapolation (ICLR 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.
