arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1506.01418stat.ML
收藏

矩阵分解模型的并行随机梯度马尔可夫链蒙特卡罗

Parallel Stochastic Gradient Markov Chain Monte Carlo for Matrix Factorisation Models

Umut Şimşekli, Hazal Koptagel, Hakan Güldaş, A. Taylan Cemgil, Figen Öztoprak, Ş. İlker Birbil

摘要:

对于大型矩阵分解问题,我们开发了一个分布式的马尔可夫链蒙特卡罗(MCMC)方法的基础上随机梯度朗之万动力学(SGLD),我们称之为并行SGLD(PSGLD)。随着数据量的增加,PSGLD具有非常有利的缩放特性,并且在计算要求方面与基于随机梯度下降的优化方法相当。PSGLD通过利用MF模型的条件独立结构以系统的方式对数据进行子采样以允许并行化和分布式计算来实现高性能。我们提供了一个收敛性证明的算法,并验证其优越的性能在各种架构,如图形处理单元,共享内存多核系统和多计算机集群。

英文摘要:

For large matrix factorisation problems, we develop a distributed Markov Chain Monte Carlo (MCMC) method based on stochastic gradient Langevin dynamics (SGLD) that we call Parallel SGLD (PSGLD). PSGLD has very favourable scaling properties with increasing data size and is comparable in terms of computational requirements to optimisation methods based on stochastic gradient descent. PSGLD achieves high performance by exploiting the conditional independence structure of the MF models to sub-sample data in a systematic manner as to allow parallelisation and distributed computation. We provide a convergence proof of the algorithm and verify its superior performance on various architectures such as Graphics Processing Units, shared memory multi-core systems and multi-computer clusters.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑