Parallel CCD++ on GPU for Matrix Factorization
Authors:
Open Access
- 4 February 2017
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 73-83
Abstract
Matrix factorization of an incomplete matrix is useful in applications such as recommender systems. Several iterative algorithms have been proposed for matrix factorization for recommender systems, including Cyclic Coordinate Descent (CCD). Recently a variant of CCD called CCD++ was developed as an attractive algorithm for parallel implementation on multicore processors. In this paper, we address the parallelization of CCD++ for a GPU. Key considerations are the reduction of data volume transferred from/to GPU global memory and minimization of intra-warp load imbalance. Starting with a base implementation, we successively improve the GPU implementation of CCD++ using loop fusion and tiling, using performance insights from hardware counter data. The resulting algorithm is shown to be faster than the best reported multicore implementation of CCD++ as well as the best reported GPU implementation of matrix factorization (using ALS, Alternating Least Squares).Keywords
Funding Information
- US National Science Foundation (1629548)
This publication has 25 references indexed in Scilit:
- Accelerating recommender systems using GPUs2015
- NOMADProceedings of the VLDB Endowment, 2014
- Parallel matrix factorization for recommender systemsKnowledge and Information Systems, 2013
- Large-scale matrix factorization with distributed stochastic gradient descent2011
- Descent Methods for Nonnegative Matrix Factorization2011
- Dual coordinate descent methods for logistic regression and maximum entropy modelsMachine Learning, 2010
- Fast Local Algorithms for Large Scale Nonnegative Matrix and Tensor FactorizationsIEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, 2009
- Large-Scale Parallel Collaborative Filtering for the Netflix Prize2008
- A dual coordinate descent method for large-scale linear SVM2008
- Nonnegative Matrix Factorization Based on Alternating Nonnegativity Constrained Least Squares and Active Set MethodSIAM Journal on Matrix Analysis and Applications, 2008