Linear Regression Using Hilbert-Space-Valued Covariates with Unknown Reproducing Kernel
By: Xinyi Li, Margaret Hoch, Michael R. Kosorok
Potential Business Impact:
Finds patterns in complex brain images.
We present a new method of linear regression based on principal components using Hilbert-space-valued covariates with unknown reproducing kernels. We develop a computationally efficient approach to estimation and derive asymptotic theory for the regression parameter estimates under mild assumptions. We demonstrate the approach in simulation studies as well as in data analysis using two-dimensional brain images as predictors.
Similar Papers
Models Parametric Analysis via Adaptive Kernel Learning
Optimization and Control
Finds best settings for computer models.
Irregular Sampling of High-Dimensional Functions in Reproducing Kernel Hilbert Spaces
Machine Learning (CS)
Makes complex math problems easier to solve.
Nonparametric learning of covariate-based Markov jump processes using RKHS techniques
Methodology
Helps predict long-term health changes better.