English

Varying-coefficient models with isotropic Gaussian process priors

Machine Learning 2015-10-15 v2 Machine Learning

Abstract

We study learning problems in which the conditional distribution of the output given the input varies as a function of additional task variables. In varying-coefficient models with Gaussian process priors, a Gaussian process generates the functional relationship between the task variables and the parameters of this conditional. Varying-coefficient models subsume hierarchical Bayesian multitask models, but also generalizations in which the conditional varies continuously, for instance, in time or space. However, Bayesian inference in varying-coefficient models is generally intractable. We show that inference for varying-coefficient models with isotropic Gaussian process priors resolves to standard inference for a Gaussian process that can be solved efficiently. MAP inference in this model resolves to multitask learning using task and instance kernels, and inference for hierarchical Bayesian multitask models can be carried out efficiently using graph-Laplacian kernels. We report on experiments for geospatial prediction.

Keywords

Cite

@article{arxiv.1508.07192,
  title  = {Varying-coefficient models with isotropic Gaussian process priors},
  author = {Matthias Bussas and Christoph Sawade and Tobias Scheffer and Niels Landwehr},
  journal= {arXiv preprint arXiv:1508.07192},
  year   = {2015}
}

Comments

17 pages, 4 Figures, minor revision

R2 v1 2026-06-22T10:43:42.170Z