Research map / Label-efficient and robust learning / Semi-supervised learning
Stochastic Consensus: Enhancing Semi-Supervised Learning with Consistency of Stochastic Classifiers
Hui Tang, Lin Sun, Kui Jia
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 consistency criterion among stochastic classifiers (ablation) | STOCO · CIFAR-10 (40 labels) | error rate: 9.25 -> 4.74 | Improved |
| add deep discriminative clustering pseudo labels (ablation) | STOCO w/o CC · CIFAR-10 (40 labels) | error rate: 11.27 -> 9.25 | Improved |
| use stochastic classifier instead of deterministic one (m=1 vs w/o CC) (ablation) | STOCO · CIFAR-10 (40 labels) | error rate: 9.25 -> 8.46 | Improved |
| increase number of stochastic classifiers from 1 to 5 (ablation) | STOCO · CIFAR-10 (40 labels) | error rate: 8.46 -> 4.74 | Improved |
| increase number of stochastic classifiers from 5 to 20 (ablation) | STOCO · CIFAR-10 (40 labels) | error rate: 4.74 -> 7.19 | Hurt |
| add stochastic consensus to FixMatch | FixMatch · CIFAR-100 (400 labels) | error rate: 48.85 -> 41.45 | Improved |
| add stochastic consensus to FixMatch | FixMatch · CIFAR-10 (40 labels) | error rate: 13.81 -> 7.17 | Improved |
| add stochastic consensus to FixMatch | FixMatch · CIFAR-10 (4000 labels) | error rate: 4.26 -> 3.86 | Improved |
| add stochastic consensus to FixMatch | FixMatch · SVHN (1000 labels) | error rate: 2.28 -> 2.38 | No clear effect |
| add stochastic consensus to FixMatch | FixMatch · STL-10 (1000 labels) | error rate: 7.98 -> 7.79 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 4 on most shared results.
- FixMatch — FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence (NeurIPS 2020) · better on 8 of 9 shared results
- FlexMatch — FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling (NeurIPS 2021) · better on 3 of 8 shared results
- CoMatch — CoMatch: Semi-Supervised Learning With Contrastive Graph Regularization (ICCV 2021) · better on 0 of 1 shared result
- ReMixMatch — ReMixMatch: Semi-Supervised Learning with Distribution Matching and Augmentation Anchoring (ICLR 2020) · 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
- RSCFed: Random Sampling Consensus Federated Semi-Supervised Learning (CVPR 2022 · similarity 0.87)
- Graph Stochastic Neural Networks for Semi-supervised Learning (NeurIPS 2020 · similarity 0.87)
- Time-Consistent Self-Supervision for Semi-Supervised Learning (ICML 2020 · similarity 0.87)
- Mind the Gap: Confidence Discrepancy Can Guide Federated Semi-Supervised Learning Across Pseudo-Mismatch (CVPR 2025 · similarity 0.87)
- Keep It on a Leash: Controllable Pseudo-label Generation Towards Realistic Long-Tailed Semi-Supervised Learning (NeurIPS 2025 · similarity 0.86)
- SimPro: A Simple Probabilistic Framework Towards Realistic Long-Tailed Semi-Supervised Learning (ICML 2024 · 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.
