Research map / Learning theory and algorithms / Matrix and tensor methods
Pivotal Estimation of Linear Discriminant Analysis in High Dimensions
Ethan X. Fang, Yajun Mei, Yuyang Shi, Qunzhi Xu, Tuo Zhao
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| use PANDA instead of AdaLDA | high-dimensional LDA · Leukemia | test error: 10.64% -> 6.93% | Improved |
| use PANDA instead of LPD | high-dimensional LDA · Leukemia | test error: 9.28% -> 6.93% | Improved |
| use PANDA instead of AdaLDA | LDA · Approximately sparse p=400 | l2 estimation error: 1.0117 -> 0.8205 | Improved |
| use PANDA instead of LPD | LDA · Block sparse (5,400) | l2 estimation error: 0.5066 -> 0.4332 | Improved |
| use PANDA instead of AdaLDA | LDA · Varying diagonal (5,400) | l2 estimation error: 0.2753 -> 0.3113 | Hurt |
| use PANDA instead of AdaLDA | LDA · Varying diagonal and AR(1) misclassification | misclassification rate: 0.0508 -> 0.0512 | No clear effect |
| increase penalty weight c from 1e-3 to 0.1 (ablation) | PANDA · AR(1) n=200 p=400 s=5 | misclassification rate: 0.3729 -> 0.2106 | Improved |
| increase penalty weight c from 0.1 to 100 (ablation) | PANDA · AR(1) n=200 p=400 s=5 | misclassification rate: 0.2106 -> 0.2035 | No clear effect |
| use PANDA instead of AdaLDA | LDA · Varying diagonal p=400 s=5 | running time: 106.739 -> 70.202 | Improved |
| use PANDA instead of AdaLDA | LDA · Varying diagonal p=800 s=5 | running time: 413.262 -> 325.486 | Improved |
| use PANDA instead of LPD | LDA · AR(1) variable selection (5,400) | true positive count: 1.95 -> 2.20 | Improved |
Closest papers by meaning
- Separability-Oriented Subclass Discriminant Analysis (TPAMI 2017 · similarity 0.86)
- The Geometry of Nonlinear Embeddings in Kernel Discriminant Analysis (TPAMI 2022 · similarity 0.85)
- Ratio Sum Versus Sum Ratio for Linear Discriminant Analysis (TPAMI 2021 · similarity 0.85)
- A Generalized Geometric Theoretical Framework of Centroid Discriminant Analysis for Linear Classification of Multi-dimensional Data (ICLR 2026 · similarity 0.84)
- Fast Locality Discriminant Analysis With Adaptive Manifold Embedding (TPAMI 2022 · similarity 0.84)
- Multi-Response Linear Discriminant Analysis in High Dimensions (JMLR 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.
