Research map / Learning theory and algorithms / Kernels and regression theory
Dimension Reduction and MARS
Yu Liu LIU, Degui Li, Yingcun Xia
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 MARS on raw covariates with SDR directions estimated from MARS gradients (drMARS) | MARS · data.3 Parkinsons, train 1000 | rMSPE (%): 32.16 -> 12.79 | Improved |
| replace MARS on raw covariates with SDR directions estimated from MARS gradients (drMARS) | MARS · data.6 Hill-Valley, train 404 | MCR (%): 19.38 -> 6.21 | Improved |
| replace MARS on raw covariates with SDR directions (drMARS) | MARS · Simulation M1, p=50, n=500, uniform X | MSE(G): 0.52 -> 0.09 | Improved |
| use drMARS instead of MARS on data.2 | MARS · data.2 house prices, train 1000 | rMSPE (%): 15.50 -> 14.62 | Improved |
| use drMARS instead of MARS on pistachio classification | MARS · data.5 pistachio, train 716 | MCR (%): 9.30 -> 9.07 | No clear effect |
| use random directions instead of estimated SDR directions (ablation) | SDR-MARS · data.1 concrete, train 343 | rMSPE (%): 15.85 -> 21.42 | Hurt |
| estimate SDR directions with pHd instead of drMARS (ablation) | SDR-MARS · data.4 residential building, train 124 | rMSPE (%): 5.85 -> 12.82 | Hurt |
| estimate SDR directions with drMARS instead of MAVE | SDR estimation · Simulation M4, X~N, p=50, n=500 | SDR error D: 0.82 -> 0.04 | Improved |
| use drMARS instead of MAVE for SDR estimation under uniform X | SDR estimation · Simulation M5, X~U, p=50, n=500 | SDR error D: 0.11 -> 0.49 | Hurt |
| use drMARS instead of CVE for SDR computing time | SDR estimation · Simulation M1, X~N, p=100, n=500 | computing time (s): 55.85 -> 6.43 | Improved |
| use drMARS instead of MAVE for choosing the true SDR dimension | SDR dimension selection · Simulation M7, X~N, p=100, n=500 | proportion selecting true dimension: 0.07 -> 0.53 | Improved |
| use drMARS instead of RF as regression estimator | RF · Simulation M1, p=50, n=500, uniform X | MSE(G): 0.61 -> 0.09 | Improved |
Closest papers by meaning
- A Unified Analysis of Multi-task Functional Linear Regression Models with Manifold Constraint and Composite Quadratic Penalty (JMLR 2023 · similarity 0.84)
- More Efficient Estimation of Multivariate Additive Models Based on Tensor Decomposition and Penalization (JMLR 2024 · similarity 0.84)
- The Maximum Separation Subspace in Sufficient Dimension Reduction with Categorical Response (JMLR 2020 · similarity 0.84)
- AKORN: Adaptive Knots generated Online for RegressioN splines (ICML 2025 · similarity 0.84)
- High-Dimensional Tensor Regression With Oracle Properties (ICML 2025 · similarity 0.83)
- Fair Multivariate Adaptive Regression Splines for Ensuring Equity and Transparency (AAAI 2024 · 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.
