Research map / Label-efficient and robust learning / Noisy labels
Deep Clustering with Incomplete Noisy Pairwise Annotations: A Geometric Regularization Approach
Read the paperCheck 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 |
|---|---|---|---|
| volume regularization (-log det(MM^T)) plus confusion-matrix loss added to logistic DCC loss | VolMaxDCC vs VanillaDCC · STL-10, CIFAR-10, ImageNet-10 with machine-annotator noise | ACC | Helped |
| volume regularization plus confusion model added to logistic DCC | VolMaxDCC vs VanillaDCC · STL-10 at 0% noise | ACC: 0.93 vs 0.91 seen | Hurt |
| volume regularization plus confusion model added to logistic DCC | VolMaxDCC vs VanillaDCC · CIFAR-10 and ImageNet-10 at 0% noise | ACC | No clear effect |
| volume regularization plus confusion model added to logistic DCC on real AMT annotations | VolMaxDCC vs C-IDEC · ImageNet-10 AMT annotations, 23.09% noise | ACC: 0.95 vs 0.91 seen | Helped |
| logistic DCC loss relies on uniform confusion assumption; no finite-sample guarantee under SSC | VolMaxDCC · theory | identifiability | No clear effect |
Closest papers by meaning
- Angular Constraint Embedding via SpherePair Loss for Constrained Clustering (NeurIPS 2025 · similarity 0.87)
- Uncertainty-Aware Probabilistic Constrained Clustering from Entangled Pairwise Supervision (NeurIPS 2026 · similarity 0.86)
- You Can Trust Your Clustering Model: A Parameter-free Self-Boosting Plug-in for Deep Clustering (NeurIPS 2025 · similarity 0.86)
- From Logits to Hierarchies: Hierarchical Clustering made Simple (ICML 2025 · similarity 0.86)
- Hard Regularization to Prevent Deep Online Clustering Collapse without Data Augmentation (AAAI 2024 · similarity 0.86)
- Interactive Deep Clustering via Value Mining (NeurIPS 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.