Research map / Learning theory and algorithms / Prediction and decision losses
Multilabel Classification with Group Testing and Codes
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 group testing (MLGT) with GT matrices instead of compressed sensing (MLCS) with regression and OMP recovery | multilabel classification with sparse labels · RCV1-2K, EURLex-4K, Delicious, AmazonCat-13K (Table 1) | Test Hamming error and Test P@k: e.g., RCV1-2K test error 2.106 (RS) vs 2.84 (Gaussian CS) | Helped |
| Use group testing with RS, expander, or sparse random matrices instead of OvsA | multilabel classification · Mediamill and Bibtex (Table 2) | Test Hamming error: Mediamill 0.68 vs 2.07; Bibtex 1.17 vs 2.55 | Helped |
| Use group testing (MLGT) instead of embedding methods (ML-CSSP, PLST, SLEEC) | multilabel classification on small datasets · Table 3 datasets | Test Hamming error: MLGT beats SLEEC on Mediamill and Bibtex but loses on Delicious (4.995 vs 4.790) | Mixed |
| Use a random 0/1 GT matrix with error-tolerant recovery (ε-tolerance) instead of exact disjunct recovery | MLGT · theory | measures needed classifiers m: m reduced from O(k^2 log d) to O(k log d) | Helped |
| Tune the error-correction parameter e at prediction time (no retraining) | MLGT · prediction | training P@k | No clear effect |
Closest papers by meaning
- Multilabel Classification by Hierarchical Partitioning and Data-dependent Grouping (NeurIPS 2020 · similarity 0.90)
- PD-Sparse : A Primal and Dual Sparse Approach to Extreme Multiclass and Multilabel Classification (ICML 2016 · similarity 0.88)
- Reliable Multilabel Classification: Prediction with Partial Abstention (AAAI 2020 · similarity 0.85)
- Taming the Sigmoid Bottleneck: Provably Argmaxable Sparse Multi-Label Classification (AAAI 2024 · similarity 0.85)
- Regret Bounds for Multilabel Classification in Sparse Label Regimes (NeurIPS 2022 · similarity 0.85)
- Consistent algorithms for multi-label classification with macro-at-$k$ metrics (ICLR 2024 · similarity 0.83)
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.