Broadcasted Nonparametric Tensor Regression
arXiv:2008.12927 · doi:10.1093/jrsssb/qkae027
Abstract
We propose a novel use of a broadcasting operation, which distributes univariate functions to all entries of the tensor covariate, to model the nonlinearity in tensor regression nonparametrically. A penalized estimation and the corresponding algorithm are proposed. Our theoretical investigation, which allows the dimensions of the tensor covariate to diverge, indicates that the proposed estimation yields a desirable convergence rate. We also provide a minimax lower bound, which characterizes the optimality of the proposed estimator for a wide range of scenarios. Numerical experiments are conducted to confirm the theoretical findings, and they show that the proposed model has advantages over its existing linear counterparts.
References in corpus (15)
- Tensor Regression with Applications in Neuroimaging Data Analysis
- Component selection and smoothing in multivariate nonparametric regression
- Variable selection in nonparametric additive models
- High-dimensional additive modeling
- Regularized Matrix Regression
- Multilinear tensor regression for longitudinal relational data
- Estimation and variable selection for generalized additive partial linear models
- Tensor-on-tensor regression
- STORE: Sparse Tensor Response Regression and Neuroimaging Analysis
- Sparse single-index model
- Estimation with Norm Regularization
- Boosted Sparse and Low-Rank Tensor Regression
- Doubly Decomposing Nonparametric Tensor Regression
- Nonparametric estimation when data on derivatives are available
- CP Degeneracy in Tensor Regression