Joshi B.,French National Center for Scientific Research |
Joshi B.,Computer Science Laboratory LIG |
Iutzeler F.,French National Center for Scientific Research |
Iutzeler F.,Applied Mathematics Laboratory LJK |
And 2 more authors.
RecSys 2016 - Proceedings of the 10th ACM Conference on Recommender Systems | Year: 2016
We introduce an asynchronous distributed stochastic gradi- ent algorithm for matrix factorization based collaborative filtering. The main idea of this approach is to distribute the user-rating matrix across different machines, each having ac- cess only to a part of the information, and to asynchronously propagate the updates of the stochastic gradient optimiza- tion across the network. Each time a machine receives a parameter vector, it averages its current parameter vector with the received one, and continues its iterations from this new point. Additionally, we introduce a similarity based reg- ularization that constrains the user and item factors to be close to the average factors of their similar users and items found on subparts of the distributed user-rating matrix. We analyze the impact of the regularization terms on Movie- Lens (100K, 1M, 10M) and NetFlix datasets and show that it leads to a more effcient matrix factorization in terms of Root Mean Square Error (RMSE) and Mean Absolute Er- ror (MAE), and that the asynchronous distributed approach significantly improves in convergence time as compared to an equivalent synchronous distributed approach. © 2016 ACM.