大规模机器学习的加速并行优化方法

Nov, 2014

大规模机器学习的加速并行优化方法

Accelerated Parallel Optimization Methods for Large Scale Machine Learning

Haipeng Luo, Patrick Haffner, Jean-Francois Paiement

TL;DR结合并行化和Nesterov加速技术，设计了更高效的算法，用于高维数据的L1正则化损失函数，包括优化BOOM和提高Shotgun的收敛速度。

Abstract

The growing amount of high dimensional data in different machine learning applications requires more efficient and scalable optimization algorithms. In this work, we consider combining two techniques,