Research map / Learning theory and algorithms / Kernels and regression theory
Nonlinear function-on-function regression by RKHS
Peijun Sang, Bing Li
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 nonlinear RKHS operator model (NLFFR) instead of linear FPCA | function-on-function regression · Simulation dense, Model 1, X from GRB, sigma 0.1 | median prediction error: 13.09 -> 3.46 | Improved |
| use NLFFR instead of linear PLFFR | function-on-function regression · Simulation dense, Model 2, X from GRB, sigma 0.1 | median prediction error: 3.01 -> 0.19 | Improved |
| use NLFFR instead of linear FPCA on real ocean data | function-on-function regression · Ocean data Temperature to Salinity | median prediction error: 1.24 × 103 -> 1.21 × 10−2 | Improved |
| use FUA instead of NLFFR | function-on-function regression · Simulation dense, Model 1, X from GRB, sigma 0.1 | median prediction error: 3.46 -> 9.53e-4 | Improved |
| use FUA instead of NLFFR | function-on-function regression · Ocean data Temperature to Salinity | median prediction error: 1.21 × 10−2 -> 1.16 × 10−2 | No clear effect |
| use FUA instead of NLFFR in sparse design | function-on-function regression · Simulation sparse design, NX in {4,5,6}, sigma 0.1 | median ISE: 17.91 -> 13.58 | Improved |
| use BMC kernel instead of GRB kernel for inner products when X generated by BMC | NLFFR · Simulation dense, Model 1, X from BMC, sigma 0.1 | median prediction error: 0.56 -> 0.43 | Improved |
| use wrong kernel (GRB) for inner products when X generated by BMC | NLFFR · Simulation dense, Model 2, X from BMC, sigma 0.1 | median prediction error: 0.45 -> 0.47 | No clear effect |
| increase noise level sigma from 0.1 to 2 | NLFFR (GRB) · Simulation dense, Model 1, X from GRB | median prediction error: 3.46 -> 3.49 | No clear effect |
| increase observations per curve from {4,5,6} to {16,18,20} | NLFFR · Simulation sparse design, sigma 0.1 | median ISE: 17.91 -> 8.15 | Improved |
Closest papers by meaning
- Functional Linear Regression with Mixed Predictors (JMLR 2022 · similarity 0.90)
- Optimal Functional Bilinear Regression with Two-dimensional Functional Covariates via Reproducing Kernel Hilbert Space (JMLR 2025 · similarity 0.88)
- Experimental Design for Linear Functionals in Reproducing Kernel Hilbert Spaces (NeurIPS 2022 · similarity 0.87)
- Operator-valued Kernels for Learning from Functional Response Data (JMLR 2016 · similarity 0.86)
- On Hypothesis Transfer Learning of Functional Linear Models (ICML 2024 · similarity 0.82)
- Bayesian Optimization on Function Spaces via Sparse RKHS Manifolds (NeurIPS 2026 · 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.
