Research map / Label-efficient and robust learning / Self-supervised and contrastive
On the Generalization and Causal Explanation in Self-Supervised Learning
Wenwen Qiang, Zeen Song, Ziyin Gu, Jiangmeng Li, Changwen Zheng, Fuchun Sun, Hui Xiong
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 UMM fine-tuning | SimSiam (ResNet-18) · CIFAR-10 linear | accuracy: 91.71 -> 92.81 | Improved |
| add UMM fine-tuning | BYOL (ResNet-18) · STL-10 linear | accuracy: 91.99 -> 93.99 | Improved |
| add UMM fine-tuning | VICRegL (ResNet-18) · CIFAR-100 linear | accuracy: 68.03 -> 69.21 | Improved |
| add UMM fine-tuning | BYOL (ResNet-50) · ImageNet-100 linear top-1 | accuracy: 75.66 -> 76.98 | Improved |
| add UMM fine-tuning | RELIC v2 (ResNet-50) · ImageNet linear, 1000 epochs | top-1 accuracy: 77.1 -> 78.2 | Improved |
| add UMM fine-tuning | SimCLR (ResNet-50) · ImageNet 1% labels semi-supervised, 1000 epochs | top-1 accuracy: 48.3 -> 52.6 | Improved |
| add UMM fine-tuning | MoCo (C4) · COCO detection | AP: 39.3 -> 40.7 | Improved |
| add coding-rate-reduction term to SSL plus JS alignment term (full UMM) (ablation) | SimSiam (ResNet-18) · CIFAR-10 linear | accuracy: 91.91 -> 92.81 | Improved |
| add JS alignment term to SSL plus coding-rate-reduction term (full UMM) (ablation) | SimSiam (ResNet-18) · CIFAR-10 linear | accuracy: 91.97 -> 92.81 | Improved |
| add coding-rate-reduction term to SSL plus JS term (full UMM) (ablation) | SimCLR (ResNet-18) · CIFAR-10 linear | accuracy: 92.03 -> 92.93 | Improved |
| add only JS term to SSL loss (ablation) | Barlow Twins (ResNet-18) · CIFAR-10 linear | accuracy: 90.88 -> 90.91 | No clear effect |
| vary coding-rate epsilon from 0.1 to 0.9 (ablation) | SimCLR + UMM · CIFAR-10 linear | accuracy: 92.93 -> 92.92 | No clear effect |
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.
- Barlow Twins — Barlow Twins: Self-Supervised Learning via Redundancy Reduction (ICML 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
- Memorization in Self-Supervised Learning Improves Downstream Generalization (ICLR 2024 · similarity 0.91)
- The Pitfalls of Memorization: When Memorization Hurts Generalization (ICLR 2025 · similarity 0.88)
- On the Over-Memorization During Natural, Robust and Catastrophic Overfitting (ICLR 2024 · similarity 0.87)
- To Grok or not to Grok: Disentangling Generalization and Memorization on Corrupted Algorithmic Datasets (ICLR 2024 · similarity 0.87)
- On the Out-of-Distribution Generalization of Self-Supervised Learning (ICML 2025 · similarity 0.86)
- Towards Memorization Estimation: Fast, Formal and Free (ICML 2025 · 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.
