Projected Newton-type methods in machine learning
An algorithmic framework for projected Newton- type methods for solving large-scale optimization problems arising in machine learning and related fields is introduced and how to apply the Newton-type framework to handle non-smooth objectives is shown.
Abstract
We consider projected Newton-type methods for solving large-scale optimization problems arising in machine learning and related fields. We first introduce an algorithmic framework for projected Newton-type methods by reviewing a canonical projected (quasi-)Newton method. This method, while conceptually pleasing, has a high computation cost per iteration. Thus, we discuss two variants that are more scalable, namely, two-metric projection and inexact projection methods. Finally, we show how to apply the Newton-type framework to handle non-smooth objectives. Examples are provided throughout the chapter to illustrate machine learning applications of our framework.