Parallel implementations for solving matrix factorization problems with optimization

Official URL: http://risc01.sabanciuniv.edu/record=b1640307 (Table of Contents)

During recent years, the exponential increase in data sets' sizes and the need for fast and accurate tools which can operate on these huge data sets in applications such as recommendation systems has led to an ever growing attention towards devising novel methods which can incorporate all the available resources to execute desired operations in the least possible time. In this work, we provide a framework for parallelized large-scale matrix factoriza- tion problems. One of the most successful and used methods to solve these problems is solving them via optimization techniques. Optimization methods require gradient vectors to update the iterates. The time spent to solve such a problem is mostly spent on calls to gradient and function value evaluations. In this work, we have used a recent method, which has not been used before for matrix factorization. When it comes to parallelization, we present both CPU and GPU implementations. As our experiments show, the proposed parallelization scales quite well. We report our results on Movie- Lens data set. Our results show that the new method is quite successful in reducing the number of iterations. We obtain very good RMSE values with signi cant promising scaling gures.