Research map / Label-efficient and robust learning / Knowledge distillation
Cluster-aware Semi-supervised Learning: Relational Knowledge Distillation Provably Learns Clustering
Yijun Dong, Kevin Miller, Qi Lei, Rachel Ward
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 relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-10, Uniform labels, N=50000, High augmentation | top-1 accuracy: 89.47 -> 89.88 | No clear effect |
| add relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-10, Uniform labels, N=50000, Weak augmentation | top-1 accuracy: 81.36 -> 84.61 | Improved |
| add relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-10, Uniform labels, N=25000, High augmentation | top-1 accuracy: 81.73 -> 87.06 | Improved |
| add relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-10, StochasticGreedy labels, N=50000, High augmentation | top-1 accuracy: 85.48 -> 91.19 | Improved |
| add relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-10, StochasticGreedy labels, N=40000, Weak augmentation | top-1 accuracy: 70.70 -> 85.12 | Improved |
| add relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-10, StochasticGreedy labels, N=25000, Medium augmentation | top-1 accuracy: 73.97 -> 82.09 | Improved |
| add relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-100, Uniform labels, N=50000, High augmentation | top-1 accuracy: 46.85 -> 49.33 | Improved |
| add relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-100, Uniform labels, N=40000, Medium augmentation | top-1 accuracy: 42.98 -> 45.78 | Improved |
| add relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-100, Uniform labels, N=25000, Weak augmentation | top-1 accuracy: 33.59 -> 35.69 | Improved |
| add relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-100, StochasticGreedy labels, N=50000, Medium augmentation | top-1 accuracy: 48.64 -> 50.67 | Improved |
| add relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-100, StochasticGreedy labels, N=40000, High augmentation | top-1 accuracy: 45.93 -> 47.62 | Improved |
| add relational knowledge distillation (RKD) loss to FixMatch | FixMatch · CIFAR-100, StochasticGreedy labels, N=25000, Weak augmentation | top-1 accuracy: 38.22 -> 39.20 | Improved |
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.
- FixMatch — FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence (NeurIPS 2020) · better on 20 of 20 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
- A Graph-Theoretic Framework for Understanding Open-World Semi-Supervised Learning (NeurIPS 2023 · similarity 0.89)
- Data driven semi-supervised learning (NeurIPS 2021 · similarity 0.88)
- A Flexible Generative Framework for Graph-based Semi-supervised Learning (NeurIPS 2019 · similarity 0.87)
- Clustering Properties of Self-Supervised Learning (ICML 2025 · similarity 0.86)
- Revisiting Self-Supervised Heterogeneous Graph Learning from Spectral Clustering Perspective (NeurIPS 2024 · similarity 0.86)
- Dynamic Sub-graph Distillation for Robust Semi-supervised Continual Learning (AAAI 2024 · 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.
