Research map / Label-efficient and robust learning / Semi-supervised learning
Contrastive Learning for Semi-Supervised Deep Regression With Generalized Ordinal Rankings From Spectral Seriation
Ce Wang, Weihang Dai, Hanru Bai, Xiaomeng Li
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 labeled samples to the ranking similarity matrix (ablation) | CLSS · IXI (1/5 labels) | MAE: 9.58 -> 9.50 | Mixed |
| add memory-based feature selection module (ablation) | CLSS · IXI (1/5 labels) | MAE: 9.58 -> 9.49 | Improved |
| add both labeled ranking and feature selection (GCLSS) (ablation) | CLSS · IXI (1/5 labels) | MAE: 9.58 -> 8.98 | Improved |
| unlabeled batch size 8 instead of 4 (ablation) | GCLSS · IXI (1/5 labels) | MAE: 10.17 -> 9.05 | Improved |
| unlabeled batch size 32 instead of 8 (ablation) | GCLSS · IXI (1/5 labels) | MAE: 9.05 -> 12.37 | Hurt |
| lambda_SC 0.1 instead of 1.0 (ablation) | GCLSS · IXI (1/5 labels) | MAE: 9.31 -> 9.05 | Improved |
| lambda_UC 0.05 instead of 0.5 (ablation) | GCLSS · IXI (1/5 labels) | MAE: 9.88 -> 9.05 | Improved |
| lambda_UR 0.01 instead of 0.001 (ablation) | GCLSS · IXI (1/5 labels) | MAE: 10.06 -> 9.05 | Improved |
| select 6 features instead of 8 in MFSM (ablation) | GCLSS · IXI (1/5 labels) | MAE: 9.05 -> 8.98 | Mixed |
| GCLSS versus CLSS | GCLSS · IXI (1/2 labels) | R2 (%): 48.9 -> 51.0 | Improved |
| GCLSS versus CLSS | GCLSS · Synthetic operator learning (1/5 labels) | MAE: 0.033 -> 0.024 | Improved |
| GCLSS versus CLSS | GCLSS · AgeDB-DIR (1/30 labels) | MAE: 9.95 -> 9.86 | 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.
- RankUp — RankUp: Boosting Semi-Supervised Regression with an Auxiliary Ranking Classifier (NeurIPS 2024) · better on 3 of 4 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
- ConR: Contrastive Regularizer for Deep Imbalanced Regression (ICLR 2024 · similarity 0.88)
- Continuous Contrastive Learning for Long-Tailed Semi-Supervised Recognition (NeurIPS 2024 · similarity 0.87)
- ACCon: Angle-Compensated Contrastive Regularizer for Deep Regression (AAAI 2025 · similarity 0.86)
- Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data (AAAI 2024 · similarity 0.86)
- Unlocking the Potential of Unlabeled Data in Semi-Supervised Domain Generalization (CVPR 2025 · similarity 0.86)
- Out-Of-Domain Unlabeled Data Improves Generalization (ICLR 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.
