Multi-Scale Methods for Machine Learning

Inderjit S. Dhillon Professor, UT Austin Abstract Many modern applications require us to very quickly find relevant results from an enormous output space of potential candidates, for example, finding the best matching product from a large catalog or suggesting related search phrases on a search engine. The size of the output space for these problems can be in the millions to billions. Moreover, observational or training data is often limited for many of the so-called “long-tail” of items in the output space. Given the inherent paucity of training data for most of the items in the output space, developing machine learning models that perform well for spaces of this size is a contemporary challenge. In this talk, I will present a multi-scale machine learning framework called Prediction for Enormous and Correlated Output Spaces (PECOS). PECOS proceeds by first building a hierarchy over the output space using unsupervised learning, and then learning a machine learning model that makes predictions at each level o