Research map / Architectures and efficiency / Neural architecture search
Learning to Auto Weight: Entirely Data-Driven and Highly Efficient Weighting Framework
Zhenmao Li, Yichao Wu, Ken Chen, Yudong Wu, Shunfeng Zhou, Jiaheng Liu, Jiaheng Liu, Junjie Yan
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 |
|---|---|---|---|
| Apply learned LAW example weighting instead of equal sample weighting | ResNet-18 · CIFAR-10-Noisy | top-1 accuracy: 79.44 -> 86.33 | Improved |
| Apply learned LAW example weighting instead of equal sample weighting | WRN-28-10 · CIFAR-100-Noisy | top-1 accuracy: 53.11 -> 68.23 | Improved |
| Use LAW rather than MentorNet | WRN-28-10 · CIFAR-10-Noisy | top-1 accuracy: 87.56 -> 89.73 | Improved |
| Use LAW rather than MentorNet | WRN-28-10 · CIFAR-100-Noisy | top-1 accuracy: 65.56 -> 68.23 | Improved |
| Apply learned LAW example weighting instead of equal sample weighting | ResNet-50 · Noisy ImageNet | top-1 accuracy: 71.9 -> 73.7 | Improved |
| Apply learned LAW example weighting instead of equal sample weighting | VGG-19 · CIFAR-10-Imbalance | top-1 accuracy: 84.88 -> 85.33 | Improved |
| Use LAW rather than L2RW | ResNet-32 · CIFAR-100-Imbalance | top-1 accuracy: 41.23 -> 43.61 | Improved |
| Increase weights for samples in minority classes | LAW · CIFAR imbalance datasets | sample weight | Improved |
| Reduce weights for corrupted-label samples | LAW · Noisy CIFAR and ImageNet | training loss gap | Improved |
Closest papers by meaning
- WeightNet: Revisiting the Design Space of Weight Networks (ECCV 2020 · similarity 0.87)
- Meta-Weight-Net: Learning an Explicit Mapping For Sample Weighting (NeurIPS 2019 · similarity 0.86)
- Learning Fast Sample Re-Weighting Without Reward Data (ICCV 2021 · similarity 0.85)
- RECAST: Reparameterized, Compact weight Adaptation for Sequential Tasks (ICLR 2025 · similarity 0.84)
- Thumb on the Scale: Optimal Loss Weighting in Last Layer Retraining (NeurIPS 2025 · similarity 0.83)
- Self-Training with Dynamic Weighting for Robust Gradual Domain Adaptation (NeurIPS 2025 · similarity 0.83)
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.
