Projected Newton-type methods in machine learning

Schmidt, M; Kim, D; Sra, S.

Datensatz

DATENSATZ AKTIONENEXPORT

Zur Ablage hinzufügen

Lokale TagsFreigabegeschichteDetailsÜbersicht

Freigegeben

Buchkapitel

Projected Newton-type methods in machine learning

MPG-Autoren

/persons/resource/persons76142

Sra, S.
Dept. Empirical Inference, Max Planck Institute for Intelligent Systems, Max Planck Society;

Externe Ressourcen

https://doc.lagout.org/science/Artificial%20Intelligence/Machine%20learning/Optimization%20for%20Machine%20Learning%20%5BSra%2C%20Nowozin%20%26%20Wright%202011-09-30%5D.pdf
(beliebiger Volltext)

Volltexte (beschränkter Zugriff)

Für Ihren IP-Bereich sind aktuell keine Volltexte freigegeben.

Volltexte (frei zugänglich)

Es sind keine frei zugänglichen Volltexte in PuRe verfügbar

Ergänzendes Material (frei zugänglich)

Es sind keine frei zugänglichen Ergänzenden Materialien verfügbar

Zitation

Schmidt, M., Kim, D., & Sra, S. (2011). Projected Newton-type methods in machine learning. In S. Sra, S. Nowozin, & S. Wright (Eds.), Optimization for Machine Learning (pp. 305-330). Cambridge, MA, USA: MIT Press.

Zitierlink: https://hdl.handle.net/11858/00-001M-0000-0013-B8DE-7

Zusammenfassung

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.