Exploiting the Structure: Stochastic Gradient Methods Using Raw Clusters

The amount of data available in the world is growing faster and bigger than our ability to deal with it. However, if we take advantage of the internal structure, data may become much smaller for machine learning purposes. In this paper we focus on one of the most fundamental machine learning tasks, empirical risk minimization (ERM), and provide faster algorithms with the help from the clustering structure of the data. We introduce a simple notion of raw clustering that can be efficiently obtained with just one pass of the data, and propose two algorithms. Our variance-reduction based algorithm ClusterSVRG introduces a new gradient estimator using the clustering information, and our accelerated algorithm ClusterACDM is built on a novel Haar transformation applied to the dual space of each cluster. Our algorithms outperform their classical counterparts both in theory and practice.