We introduce a Gaussian process model of functions which are additive. An additive function is one which decomposes into a sum of low-dimensional functions, each depending on only a subset of the input variables. Additive GPs generalize both Generalized Additive Models, and the standard GP models which use squared-exponential kernels. Hyperparameter learning in this model can be seen as Bayesian Hierarchical Kernel Learning (HKL). We introduce an expressive but tractable parameterization of the kernel function, which allows efficient evaluation of all input interaction terms, whose number is exponential in the input dimension. The additional structure discoverable by this model results in increased interpretability, as well as state-of-the-art predictive power in regression tasks.
| Author(s): | Duvenaud, D. and Nickisch, H. and Rasmussen, CA. |
| Book Title: | Advances in Neural Information Processing Systems 24 |
| Pages: | 226-234 |
| Year: | 2011 |
| Day: | 0 |
| Editors: | J Shawe-Taylor and RS Zemel and P Bartlett and F Pereira and KQ Weinberger |
| Bibtex Type: | Conference Paper (inproceedings) |
| Event Name: | Twenty-Fifth Annual Conference on Neural Information Processing Systems (NIPS 2011) |
| Event Place: | Granada, Spain |
| Digital: | 0 |
| Electronic Archiving: | grant_archive |
| Links: | |
BibTex
@inproceedings{DuvenaudNR2012,
title = {Additive Gaussian Processes},
booktitle = {Advances in Neural Information Processing Systems 24},
abstract = {We introduce a Gaussian process model of functions which are additive. An additive function is one which decomposes into a sum of low-dimensional functions, each depending on only a subset of the input variables. Additive GPs generalize both Generalized Additive Models, and the standard GP models which use squared-exponential kernels. Hyperparameter learning in this model can be seen as Bayesian Hierarchical Kernel Learning (HKL). We introduce an expressive but tractable parameterization of the kernel function, which allows efficient evaluation of all input interaction terms, whose number is exponential in the input dimension. The additional structure discoverable by this model results in increased interpretability, as well as state-of-the-art predictive power in regression tasks.},
pages = {226-234},
editors = {J Shawe-Taylor and RS Zemel and P Bartlett and F Pereira and KQ Weinberger},
year = {2011},
slug = {duvenaudnr2012},
author = {Duvenaud, D. and Nickisch, H. and Rasmussen, CA.}
}