Research map / Probabilistic and scientific modeling / Molecules and proteins
Latent Discriminant Analysis with Representative Feature Discovery
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 |
|---|---|---|---|
| add data-driven GMM prior over clusters (JLFDA vs LFDA) (ablation) | latent Fisher discriminant analysis · MUSK1 | accuracy: 81.4 -> 87.1 | Helped |
| add data-driven prior (JLFDA vs LFDA) (ablation) | latent Fisher discriminant analysis · MUSK2 | accuracy: 76.4 -> 81.3 | Helped |
| add data-driven prior (JLFDA vs LFDA) (ablation) | latent Fisher discriminant analysis · Elephant | accuracy: 74.5 -> 82.2 | Helped |
| add data-driven prior (JLFDA vs LFDA) (ablation) | latent Fisher discriminant analysis · Fox | accuracy: 61.5 -> 59.5 | Hurt |
| add data-driven prior (JLFDA vs LFDA) (ablation) | latent Fisher discriminant analysis · Tiger | accuracy: 74.0 -> 80.5 | Helped |
| add data-driven prior (JLFDA vs LFDA) (ablation) | latent Fisher discriminant analysis · MUSK and Corel average | accuracy: 73.56 -> 78.1 | Helped |
| add latent cluster-level labels (LFDA vs plain LDA with bag labels) | LDA · MUSK and Corel average | accuracy: 63.54 -> 73.56 | Helped |
| use JLFDA instead of MI-SVM | JLFDA · MUSK and Corel average | accuracy: 77.08 -> 78.1 | Helped |
| use JLFDA instead of MI-SVM | JLFDA · MUSK2 and Tiger | accuracy: 84.3 -> 81.3 | Hurt |
| use JLFDA instead of SVM baselines for semantic keyframes | JLFDA · TRECVID MED11 E001-E005, No.1 counts | videos ranked first: 3 -> 7 on E005 and E002; E004 lower | Mixed |
| use JLFDA instead of SVM(2) on wedding ceremony event | JLFDA · TRECVID MED11 E004 | videos ranked first: 3 vs 7 | Hurt |
| use JLFDA instead of LFDA (ablation) | JLFDA · TRECVID MED11 E005 | videos ranked first: 1 -> 7 | Helped |
Closest papers by meaning
- Understanding Self-Supervised Learning via Latent Distribution Matching (ICML 2026 · similarity 0.85)
- Discriminative Topic Modeling with Logistic LDA (NeurIPS 2019 · similarity 0.84)
- Learning Low-Dimensional Temporal Representations (ICML 2018 · similarity 0.84)
- Feature Selection for Latent Factor Models (CVPR 2025 · similarity 0.82)
- Identifying Metric Structures of Deep Latent Variable Models (ICML 2025 · similarity 0.81)
- Cross-Entropy Is All You Need To Invert the Data Generating Process (ICLR 2025 · similarity 0.81)
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.