Research map / Learning theory and algorithms / Matrix and tensor methods
Crowdsourcing via Annotator Co-occurrence Imputation and Provable Symmetric Nonnegative Matrix Factorization
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 |
|---|---|---|---|
| Use robust co-occurrence imputation rather than equal weights (ablation) | RobSymNMF on TREC · TREC | classification error (%): 33.23 -> 30.68 | Helped |
| Use robust co-occurrence imputation rather than equal weights (ablation) | RobSymNMF on RTE · RTE | classification error (%): 7.37 -> 7.25 | Helped |
| Remove co-occurrence imputation (ablation) | SymNMF on RTE · RTE | classification error (%): 13.87 -> 48.75 | Hurt |
| Remove co-occurrence imputation (ablation) | SymNMF on TREC · TREC | classification error (%): 29.36 -> 35.47 | Hurt |
| Initialize EM with RobSymNMF (ablation) | RobSymNMF on LabelMe · LabelMe | classification error (%): 32.10 -> 22.10 | Helped |
| Use a shifted ReLU update instead of ordinary ReLU (ablation) | SymNMF · Synthetic SymNMF convergence | convergence speed | Helped |
| Use pairwise co-occurrence-based SymNMF rather than CNMF | Crowdsourcing methods on Car with 70% missing co-occurrences · Car, Miss=70% | classification error (%): 43.65 -> 24.01 | Helped |
Closest papers by meaning
- Coupled Confusion Correction: Learning from Crowds with Sparse Annotations (AAAI 2024 · similarity 0.87)
- Noisy Label Learning with Instance-Dependent Outliers: Identifiability via Crowd Wisdom (NeurIPS 2024 · similarity 0.87)
- Unbiased Multi-Label Learning from Crowdsourced Annotations (ICML 2024 · similarity 0.87)
- TLLC: Transfer Learning-based Label Completion for Crowdsourcing (ICML 2025 · similarity 0.86)
- Efficient Online Crowdsourcing with Complex Annotations (AAAI 2024 · similarity 0.85)
- KFNN: K-Free Nearest Neighbor For Crowdsourcing (NeurIPS 2024 · similarity 0.84)
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.