direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

All Publications

Regression with Linear Factored Functions
Citation key Boehmer15b
Author Böhmer, W. and Obermayer, K.
Title of Book Machine Learning and Knowledge Discovery in Databases
Pages 119-134
Year 2015
ISBN 978-3-319-23527-1, 978-3-319-23528-8
ISSN 0302-9743
DOI 10.1007/978-3-319-23528-8_8
Volume 9284
Publisher Springer International Publishing
Series Lecture Notes in Computer Science
Abstract Many applications that use empirically estimated functions face a curse of dimensionality, because integrals over most function classes must be approximated by sampling. This paper introduces a novel regression-algorithm that learns linear factored functions (LFF). This class of functions has structural properties that allow to analytically solve certain integrals and to calculate point-wise products. Applications like belief propagation and reinforcement learning can exploit these properties to break the curse and speed up computation. We derive a regularized greedy optimization scheme, that learns factored basis functions during training. The novel regression algorithm performs competitively to Gaussian processes on benchmark tasks, and the learned LFF functions are with 4-9 factored basis functions on average very compact.
Link to publication Download Bibtex entry

Zusatzinformationen / Extras

Quick Access:

Schnellnavigation zur Seite über Nummerneingabe

Auxiliary Functions