Research map / Label-efficient and robust learning / Recognition and long-tail
Enhancing Sharpness-Aware Optimization Through Variance Suppression
Bingcong Li, Georgios B. Giannakis
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 SAM with VaSSO (variance-suppressed adversary) | SAM · CIFAR10 VGG-11-BN | test accuracy (%): 93.82 -> 94.10 | Improved |
| replace SAM with VaSSO | SAM · CIFAR10 PyramidNet-110 | test accuracy (%): 97.85 -> 97.93 | Improved |
| replace SAM with VaSSO | SAM · CIFAR100 ResNet-18 | test accuracy (%): 80.96 -> 81.30 | Improved |
| replace SAM with VaSSO | SAM · CIFAR100 PyramidNet-110 | test accuracy (%): 85.60 -> 85.85 | Improved |
| replace SAM with VaSSO | SAM · ImageNet ResNet-50 | test accuracy (%): 77.16 -> 77.42 | Improved |
| replace SAM with VaSSO | SAM · ImageNet ViT-S/32 | test accuracy (%): 68.98 -> 69.54 | Improved |
| add VaSSO variance suppression to GSAM (V+G) | GSAM · ImageNet ResNet-50 | test accuracy (%): 77.20 -> 77.48 | Improved |
| add VaSSO variance suppression to GSAM (V+G) | GSAM · ImageNet ViT-S/32 | test accuracy (%): 69.42 -> 69.61 | Improved |
| replace SAM with VaSSO (theta=0.9) | SAM · IWSLT-14 De-En Transformer | validation perplexity: 5.00 -> 5.00 | No clear effect |
| replace AdamW with VaSSO (theta=0.4) | AdamW · IWSLT-14 De-En Transformer | BLEU: 34.66 -> 34.88 | Improved |
| use smaller theta (0.4 instead of 0.9) in VaSSO (ablation) | VaSSO · IWSLT-14 De-En Transformer | BLEU: 34.81 -> 34.88 | Improved |
| replace SAM with VaSSO (theta=0.2) under label noise | SAM · CIFAR10 25% label noise | test accuracy (%): 96.39 -> 96.48 | Improved |
Closest papers by meaning
- Friendly Sharpness-Aware Minimization (CVPR 2024 · similarity 0.89)
- Sharpness-Aware Minimization: General Analysis and Improved Rates (ICLR 2025 · similarity 0.88)
- Sharpness-Aware Minimization Efficiently Selects Flatter Minima Late In Training (ICLR 2025 · similarity 0.88)
- A Universal Class of Sharpness-Aware Minimization Algorithms (ICML 2024 · similarity 0.87)
- Sharpness-Aware Minimization Enhances Feature Quality via Balanced Learning (ICLR 2024 · similarity 0.87)
- Fundamental Convergence Analysis of Sharpness-Aware Minimization (NeurIPS 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.
