General Functional Matrix Factorization Using Gradient Boosting

Tianqi Chen, Hang Li, Qiang Yang, Yong Yu
; Proceedings of the 30th International Conference on Machine Learning, PMLR 28(1):436-444, 2013.

Abstract

Matrix factorization is among the most successful techniques for collaborative filtering. One challenge of collaborative filtering is how to utilize available auxiliary information to improve prediction accuracy. In this paper, we study the problem of utilizing auxiliary information as features of factorization and propose formalizing the problem as general functional matrix factorization, whose model includes conventional matrix factorization models as its special cases. Moreover, we propose a gradient boosting based algorithm to efficiently solve the optimization problem. Finally, we give two specific algorithms for efficient feature function construction for two specific tasks. Our method can construct more suitable feature functions by searching in an infinite functional space based on training data and thus can yield better prediction accuracy. The experimental results demonstrate that the proposed method outperforms the baseline methods on three real-world datasets.

Cite this Paper


BibTeX
@InProceedings{pmlr-v28-chen13e, title = {General Functional Matrix Factorization Using Gradient Boosting}, author = {Tianqi Chen and Hang Li and Qiang Yang and Yong Yu}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {436--444}, year = {2013}, editor = {Sanjoy Dasgupta and David McAllester}, volume = {28}, number = {1}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/chen13e.pdf}, url = {http://proceedings.mlr.press/v28/chen13e.html}, abstract = {Matrix factorization is among the most successful techniques for collaborative filtering. One challenge of collaborative filtering is how to utilize available auxiliary information to improve prediction accuracy. In this paper, we study the problem of utilizing auxiliary information as features of factorization and propose formalizing the problem as general functional matrix factorization, whose model includes conventional matrix factorization models as its special cases. Moreover, we propose a gradient boosting based algorithm to efficiently solve the optimization problem. Finally, we give two specific algorithms for efficient feature function construction for two specific tasks. Our method can construct more suitable feature functions by searching in an infinite functional space based on training data and thus can yield better prediction accuracy. The experimental results demonstrate that the proposed method outperforms the baseline methods on three real-world datasets.} }
Endnote
%0 Conference Paper %T General Functional Matrix Factorization Using Gradient Boosting %A Tianqi Chen %A Hang Li %A Qiang Yang %A Yong Yu %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-chen13e %I PMLR %J Proceedings of Machine Learning Research %P 436--444 %U http://proceedings.mlr.press %V 28 %N 1 %W PMLR %X Matrix factorization is among the most successful techniques for collaborative filtering. One challenge of collaborative filtering is how to utilize available auxiliary information to improve prediction accuracy. In this paper, we study the problem of utilizing auxiliary information as features of factorization and propose formalizing the problem as general functional matrix factorization, whose model includes conventional matrix factorization models as its special cases. Moreover, we propose a gradient boosting based algorithm to efficiently solve the optimization problem. Finally, we give two specific algorithms for efficient feature function construction for two specific tasks. Our method can construct more suitable feature functions by searching in an infinite functional space based on training data and thus can yield better prediction accuracy. The experimental results demonstrate that the proposed method outperforms the baseline methods on three real-world datasets.
RIS
TY - CPAPER TI - General Functional Matrix Factorization Using Gradient Boosting AU - Tianqi Chen AU - Hang Li AU - Qiang Yang AU - Yong Yu BT - Proceedings of the 30th International Conference on Machine Learning PY - 2013/02/13 DA - 2013/02/13 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-chen13e PB - PMLR SP - 436 DP - PMLR EP - 444 L1 - http://proceedings.mlr.press/v28/chen13e.pdf UR - http://proceedings.mlr.press/v28/chen13e.html AB - Matrix factorization is among the most successful techniques for collaborative filtering. One challenge of collaborative filtering is how to utilize available auxiliary information to improve prediction accuracy. In this paper, we study the problem of utilizing auxiliary information as features of factorization and propose formalizing the problem as general functional matrix factorization, whose model includes conventional matrix factorization models as its special cases. Moreover, we propose a gradient boosting based algorithm to efficiently solve the optimization problem. Finally, we give two specific algorithms for efficient feature function construction for two specific tasks. Our method can construct more suitable feature functions by searching in an infinite functional space based on training data and thus can yield better prediction accuracy. The experimental results demonstrate that the proposed method outperforms the baseline methods on three real-world datasets. ER -
APA
Chen, T., Li, H., Yang, Q. & Yu, Y.. (2013). General Functional Matrix Factorization Using Gradient Boosting. Proceedings of the 30th International Conference on Machine Learning, in PMLR 28(1):436-444

Related Material