Parallel Dual Coordinate Descent Method for Large-scale Linear Classification in Multi-core Environments
Wei-Lin Chiang, National Taiwan University; Mu-Chu Lee, National Taiwan University; Chih-Jen Lin*, National Taiwan University
Dual coordinate descent method is one of the most eﬀective approaches for large-scale linear classiﬁcation. However, its sequential design makes the parallelization diﬃcult. In this work, we target at the parallelization in a multi-core environment. After pointing out diﬃculties faced in some existing approaches, we propose a new framework to parallelize the dual coordinate descent method. The key idea is to make the majority of all operations (gradient calculation here) parallelizable. The proposed framework is shown to be theoretically sound. Further, we demonstrate through experiments that the new framework is robust and eﬃcient in a multi-core environment.