Research map / Label-efficient and robust learning / Self-supervised and contrastive
Unsupervised Visual Representation Learning by Synchronous Momentum Grouping
Bo Pang, Yifan Zhang, Yaoyi Li, Jia Cai, Cewu Lu
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 |
|---|---|---|---|
| use linearly decreasing beta schedule 1.0->0.99 instead of fixed beta 0.99 (ablation) | SMoG · ImageNet linear (100 ep) | top-1 accuracy: 65.9 -> 67.2 | Improved |
| vary final beta of schedule (0.9, 0.99, 0.999) (ablation) | SMoG · ImageNet linear (100 ep) | top-1 accuracy: 67.0 / 67.2 / 67.1 | No clear effect |
| add periodical clustering to avoid collapse (ablation) | SMoG · ImageNet linear (100 ep) | top-1 accuracy: 0.1 -> 53.7 | Improved |
| add periodical reset of the momentum network on top of periodical clustering (ablation) | SMoG · ImageNet linear (100 ep) | top-1 accuracy: 53.7 -> 67.2 | Improved |
| reset momentum network periodically without clustering (ablation) | SMoG · ImageNet linear (100 ep) | top-1 accuracy: 0.1 -> 53.4 | Improved |
| increase number of groups from 300 to 3000 (ablation) | SMoG · ImageNet linear (100 ep) | top-1 accuracy: 65.2 -> 67.2 | Improved |
| increase number of groups from 3000 to 10000 (ablation) | SMoG · ImageNet linear (100 ep) | top-1 accuracy: 67.2 -> 67.2 | No clear effect |
| momentum update of group features instead of averaging update (ablation) | SMoG · ImageNet linear (100 ep) | top-1 accuracy: 65.8 -> 67.2 | Improved |
| momentum update instead of adopting latest instance features as group features (ablation) | SMoG · ImageNet linear (100 ep) | top-1 accuracy: 42.1 -> 67.2 | Improved |
| momentum update instead of randomly selected group features (ablation) | SMoG · ImageNet linear (100 ep) | top-1 accuracy: 30.2 -> 67.2 | Improved |
| add multi-crop training | SMoG ResNet-50 · ImageNet linear eval | top-1 accuracy: 73.6 -> 76.4 | Improved |
| add multi-crop training | SMoG Swin-T · ImageNet linear eval | top-1 accuracy: 74.5 -> 77.7 | Improved |
Closest papers by meaning
- Momentum Contrast for Unsupervised Visual Representation Learning (CVPR 2020 · similarity 0.87)
- Self-Supervised Visual Representation Learning from Hierarchical Grouping (NeurIPS 2020 · similarity 0.87)
- Self-Supervised Visual Representation Learning with Semantic Grouping (NeurIPS 2022 · similarity 0.86)
- Jigsaw Clustering for Unsupervised Visual Representation Learning (CVPR 2021 · similarity 0.86)
- Propagate Yourself: Exploring Pixel-Level Consistency for Unsupervised Visual Representation Learning (CVPR 2021 · similarity 0.86)
- Self-Organizing Visual Prototypes for Non-Parametric Representation Learning (ICML 2025 · similarity 0.85)
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.
