Research map / Federated learning and privacy / Personalized and heterogeneous FL
RSCFed: Random Sampling Consensus Federated Semi-Supervised Learning
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 RSCFed aggregation instead of Fed-Consist | FSSL with 1 labeled and 9 unlabeled clients · SVHN | Accuracy: 66.94 -> 70.26 | Helped |
| Use RSCFed aggregation instead of FedIRM | FSSL with 1 labeled and 9 unlabeled clients · CIFAR-100 | Accuracy: 14.11 -> 15.82 | Helped |
| Use RSCFed aggregation instead of Fed-Consist | FSSL with 1 labeled and 9 unlabeled clients · ISIC 2018 | Accuracy: 68.74 -> 70.26 | Helped |
| Use RSCFed instead of Fed-Consist | Partially labeled clients (10% labeled images) · SVHN | Accuracy: 77.54 -> 79.01 | Helped |
| Add multiple random sub-sampling operations (SSO) (ablation) | Basic FSSL aggregation · SVHN | Accuracy: 66.94 -> 69.15 | Helped |
| Add multiple random sub-sampling operations (SSO) (ablation) | Basic FSSL aggregation · CIFAR-100 | AUC: 78.3 -> 81.8 | Helped |
| Add multiple random sub-sampling operations (SSO) (ablation) | Basic FSSL aggregation · ISIC 2018 | Accuracy: 68.74 -> 69.85 | Helped |
| Add distance-reweighted model aggregation (DMA) after SSO (ablation) | RSCFed · SVHN | Accuracy: 69.15 -> 70.26 | Helped |
| Add distance-reweighted model aggregation (DMA) after SSO (ablation) | RSCFed · CIFAR-100 | AUC: 81.8 -> 81.4 | Hurt |
| Add distance-reweighted model aggregation (DMA) after SSO (ablation) | RSCFed · ISIC 2018 | Accuracy: 69.85 -> 70.26 | Helped |
| Increase unlabeled clients from 4 to 49 with one labeled client (ablation) | RSCFed versus Fed-Consist · SVHN | Accuracy improvement over Fed-Consist: 1.51% to 3.98% | Helped |
| Use 0.8× Fed-Consist communication cost (ablation) | RSCFed · SVHN | Accuracy: 66.94 -> 68.23 | Helped |
Closest papers by meaning
- SemiDFL: A Semi-Supervised Paradigm for Decentralized Federated Learning (AAAI 2025 · similarity 0.90)
- Mind the Gap: Confidence Discrepancy Can Guide Federated Semi-Supervised Learning Across Pseudo-Mismatch (CVPR 2025 · similarity 0.90)
- (FL)$^2$: Overcoming Few Labels in Federated Semi-Supervised Learning (NeurIPS 2024 · similarity 0.90)
- FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning (NeurIPS 2025 · similarity 0.88)
- Combating Data Imbalances in Federated Semi-supervised Learning with Dual Regulators (AAAI 2024 · similarity 0.88)
- Progressive Distribution Matching for Federated Semi-Supervised Learning (AAAI 2025 · similarity 0.88)
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.