Research map / Learning theory and algorithms / Kernels and regression theory
MFPCA: Multiscale Functional Principal Component Analysis
Zhenhua Lin, Hongtu Zhu
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 |
|---|---|---|---|
| replace single-scale FPCA scores with multiscale FPCA scores, K=6 | random forest classifier · DTI fractional anisotropy, MS status classification | correct classification rate (%): 70.9 -> 73.4 | Improved |
| multiscale vs single-scale FPCA, K=7 | random forest classifier · DTI fractional anisotropy, MS status classification | correct classification rate (%): 71.1 -> 74.5 | Improved |
| multiscale vs single-scale FPCA, K=1 | random forest classifier · DTI fractional anisotropy, MS status classification | correct classification rate (%): 64.7 -> 66.2 | Improved |
| multiscale vs single-scale FPCA, K=3 | random forest classifier · DTI fractional anisotropy, MS status classification | correct classification rate (%): 70.5 -> 69.9 | Hurt |
| multiscale vs single-scale FPCA, K=4 | random forest classifier · DTI fractional anisotropy, MS status classification | correct classification rate (%): 69.8 -> 70.5 | No clear effect |
| use multiscale FPCA instead of wavelet (best wavelet db2), K=12 | random forest classifier · DTI fractional anisotropy, MS status classification | correct classification rate (%): 70.8 -> 72.6 | Improved |
| add more single-scale principal components (K=1 to 12) | single-scale FPCA · DTI fractional anisotropy, MS status classification | correct classification rate (%): 64.7 -> 69.9 | No clear effect |
| add more multiscale principal components beyond K=5 | multiscale FPCA · DTI fractional anisotropy, MS status classification | correct classification rate (%): 72.9 -> 74.5 | Improved |
| multiscale vs single-scale FPCA estimation on simulated data (n=50, n=200) | eigenfunction/projection estimators · simulation | log10 RMSE | Improved |
Closest papers by meaning
- Integrated Principal Components Analysis (JMLR 2021 · similarity 0.86)
- Contrastive Functional Principal Component Analysis (AAAI 2025 · similarity 0.85)
- Unconstrained Stochastic CCA: Unifying Multiview and Self-Supervised Learning (ICLR 2024 · similarity 0.83)
- Heterogeneous Sufficient Dimension Reduction and Subspace Clustering (ICML 2025 · similarity 0.83)
- Probabilistic Geometric Principal Component Analysis with application to neural data (ICLR 2025 · similarity 0.82)
- Shape-Informed Clustering of Multi-Dimensional Functional Data via Deep Functional Autoencoders (NeurIPS 2025 · similarity 0.82)
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.
