Frame-based Data Factorizations

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Authors

Archetypal Analysis is the method of choice to compute interpretable matrix factorizations. Every data point is represented as a convex combination of factors, i.e., points on the boundary of the convex hull of the data. This renders computation inefficient. In this paper, we show that the set of vertices of a convex hull, the so-called frame, can be efficiently computed by a quadratic program. We provide theoretical and empirical results for our proposed approach and make use of the frame to accelerate Archetypal Analysis. The novel method yields similar reconstruction errors as baseline competitors but is much faster to compute.
Original languageEnglish
Title of host publication34th International Conference on Machine Learning, ICML 2017
EditorsDoina Precup, Yee Whye Teh
Number of pages9
Place of PublicationRed Hook
PublisherCurran Associates
Publication date25.07.2017
Pages2305-2313
ISBN (electronic)978-1-5108-5514-4
Publication statusPublished - 25.07.2017
EventInternational Conference on Machine Learning - ICML 2017: Thirty-fourth International Conference on Machine Learning - International Convention Centre, Sydney , Sydney, Australia
Duration: 06.08.201711.08.2017
Conference number: 34
https://icml.cc/Conferences/2017

Bibliographical note

This work has been funded in parts by the German Federal
Ministry of Education and Science BMBF under grant
QQM/01LSA1503C