BriefGPT.xyz
Feb, 2015
分布式原-对偶优化中的加法与平均法
Adding vs. Averaging in Distributed Primal-Dual Optimization
HTML
PDF
Chenxin Ma, Virginia Smith, Martin Jaggi, Michael I. Jordan, Peter Richtárik...
TL;DR
本文介绍了一种分布式优化方法——CoCoA+，解决了大规模机器学习中通信瓶颈和全局参数的更新问题，使得非光滑的凸损失函数以更快的速度达到更优的性能表现。
Abstract
distributed optimization
algorithms for large-scale
machine learning
suffer from a communication bottleneck. Reducing communication makes the efficient aggregation of partial work from different machines more cha
→