Non-Linear Matrix Factorization with Gaussian Processes

Neil D. LawrenceRaquel Urtasun
,  26, 2009.

Abstract

A popular approach to collaborative filtering is matrix factorization. In this paper we develop a non-linear probabilistic matrix factorization using Gaussian process latent variable models. We use stochastic gradient descent (SGD) to optimize the model. SGD allows us to apply Gaussian processes to data sets with millions of observations without approximate methods. We apply our approach to benchmark movie recommender data sets. The results show better than previous state-of-the-art performance.

Cite this Paper


BibTeX
@InProceedings{pmlr-v-lawrence-nlmf09, title = {Non-Linear Matrix Factorization with Gaussian Processes}, author = {Neil D. Lawrence and Raquel Urtasun}, year = {}, editor = {}, volume = {26}, address = {San Francisco, CA}, url = {http://inverseprobability.com/publications/lawrence-nlmf09.html}, abstract = {A popular approach to collaborative filtering is matrix factorization. In this paper we develop a non-linear probabilistic matrix factorization using Gaussian process latent variable models. We use stochastic gradient descent (SGD) to optimize the model. SGD allows us to apply Gaussian processes to data sets with millions of observations without approximate methods. We apply our approach to benchmark movie recommender data sets. The results show better than previous state-of-the-art performance.} }
Endnote
%0 Conference Paper %T Non-Linear Matrix Factorization with Gaussian Processes %A Neil D. Lawrence %A Raquel Urtasun %B %C Proceedings of Machine Learning Research %D %E %F pmlr-v-lawrence-nlmf09 %I PMLR %J Proceedings of Machine Learning Research %P -- %U http://inverseprobability.com %V %W PMLR %X A popular approach to collaborative filtering is matrix factorization. In this paper we develop a non-linear probabilistic matrix factorization using Gaussian process latent variable models. We use stochastic gradient descent (SGD) to optimize the model. SGD allows us to apply Gaussian processes to data sets with millions of observations without approximate methods. We apply our approach to benchmark movie recommender data sets. The results show better than previous state-of-the-art performance.
RIS
TY - CPAPER TI - Non-Linear Matrix Factorization with Gaussian Processes AU - Neil D. Lawrence AU - Raquel Urtasun BT - PY - DA - ED - ID - pmlr-v-lawrence-nlmf09 PB - PMLR SP - DP - PMLR EP - L1 - UR - http://inverseprobability.com/publications/lawrence-nlmf09.html AB - A popular approach to collaborative filtering is matrix factorization. In this paper we develop a non-linear probabilistic matrix factorization using Gaussian process latent variable models. We use stochastic gradient descent (SGD) to optimize the model. SGD allows us to apply Gaussian processes to data sets with millions of observations without approximate methods. We apply our approach to benchmark movie recommender data sets. The results show better than previous state-of-the-art performance. ER -
APA
Lawrence, N.D. & Urtasun, R.. (). Non-Linear Matrix Factorization with Gaussian Processes. , in PMLR :-

Related Material