We introduce a new approach to non-linear regression called function factorization, that is suitable for problems where an output variable can reasonably be modeled by a number of multiplicative interaction terms between non-linear functions of the inputs. The idea is to approximate a complicated function on a high-dimensional space by the sum of products of simpler functions on lower-dimensional subspaces. Function factorization can be seen as a generalization of matrix and tensor factorization methods, in which the data are approximated by the sum of outer products of vectors. We present a non-parametric Bayesian approach to function factorization where the priors over the factorizing functions are warped Gaussian processes, and we do inference using Hamiltonian Markov chain Monte Carlo. We demonstrate the superior predictive performance of the method on a food science data set compared to Gaussian process regression and tensor factorization using PARAFAC and GEMANOVA models.
[1]
S. P. Smith.
Differentiation of the Cholesky Algorithm
,
1995
.
[2]
Rasmus Bro,et al.
The N-way Toolbox for MATLAB
,
2000
.
[3]
Carl E. Rasmussen,et al.
Gaussian processes for machine learning
,
2005,
Adaptive computation and machine learning.
[4]
Carl E. Rasmussen,et al.
Warped Gaussian Processes
,
2003,
NIPS.
[5]
Ryan P. Adams,et al.
Gaussian process product models for nonparametric nonstationarity
,
2008,
ICML '08.
[6]
S. Duane,et al.
Hybrid Monte Carlo
,
1987
.
[7]
Rasmus Bro,et al.
Exploring complex interactions in designed data using GEMANOVA. Color changes in fresh beef during storage
,
2002
.
[8]
R. Bro,et al.
PARAFAC and missing values
,
2005
.
[9]
Mikkel N. Schmidt,et al.
Nonnegative Matrix Factorization with Gaussian Process Priors
,
2008,
Comput. Intell. Neurosci..