Coresets for Archetypal Analysis

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Standard

Coresets for Archetypal Analysis. / Mair, Sebastian; Brefeld, Ulf.
32rd Conference on Neural Information Processing Systems (NeurIPS 2019): Vancouver, Canada, 8-14 December 2019. Hrsg. / Hanna Wallach; Hugo Larochelle. Band 10 Red Hook: Curran Associates, 2020. S. 7215-7223 (Advances in neural information processing systems; Band 32).

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Harvard

Mair, S & Brefeld, U 2020, Coresets for Archetypal Analysis. in H Wallach & H Larochelle (Hrsg.), 32rd Conference on Neural Information Processing Systems (NeurIPS 2019): Vancouver, Canada, 8-14 December 2019. Bd. 10, Advances in neural information processing systems, Bd. 32, Curran Associates, Red Hook, S. 7215-7223, 33rd Conference on Neural Information Processing Systems - NeurIPS 2019, Vancouver, British Columbia, Kanada, 08.12.19. <https://papers.nips.cc/paper/8945-coresets-for-archetypal-analysis.pdf>

APA

Mair, S., & Brefeld, U. (2020). Coresets for Archetypal Analysis. In H. Wallach, & H. Larochelle (Hrsg.), 32rd Conference on Neural Information Processing Systems (NeurIPS 2019): Vancouver, Canada, 8-14 December 2019 (Band 10, S. 7215-7223). (Advances in neural information processing systems; Band 32). Curran Associates. https://papers.nips.cc/paper/8945-coresets-for-archetypal-analysis.pdf

Vancouver

Mair S, Brefeld U. Coresets for Archetypal Analysis. in Wallach H, Larochelle H, Hrsg., 32rd Conference on Neural Information Processing Systems (NeurIPS 2019): Vancouver, Canada, 8-14 December 2019. Band 10. Red Hook: Curran Associates. 2020. S. 7215-7223. (Advances in neural information processing systems).

Bibtex

@inbook{5bfd7a9c5e8842acab0136cc700f81e5,
title = "Coresets for Archetypal Analysis",
abstract = "Archetypal analysis represents instances as linear mixtures of prototypes (the archetypes) that lie on the boundary of the convex hull of the data. Archetypes are thus often better interpretable than factors computed by other matrix factorization techniques. However, the interpretability comes with high computational cost due to additional convexity-preserving constraints. In this paper, we propose efficient coresets for archetypal analysis. Theoretical guarantees are derived by showing that quantization errors of k-means upper bound archetypal analysis; the computation of a provable absolute-coreset can be performed in only two passes over the data. Empirically, we show that the coresets lead to improved performance on several data sets.",
keywords = "Business informatics",
author = "Sebastian Mair and Ulf Brefeld",
note = "Richtige Z{\"a}hlung der Konferenz: 33rd Conference on Neural Information Processing Systems. Copyright{\textcopyright}(2019) by individual authors and Neural Information Processing Systems Foundation Inc. Printed with permission by Curran Associates, Inc. (2020); 33rd Conference on Neural Information Processing Systems - NeurIPS 2019, NeurIPS 2019 ; Conference date: 08-12-2019 Through 14-12-2019",
year = "2020",
language = "English",
isbn = "978-1-71380-793-3",
volume = "10",
series = "Advances in neural information processing systems",
publisher = "Curran Associates",
pages = "7215--7223",
editor = "Hanna Wallach and Hugo Larochelle",
booktitle = "32rd Conference on Neural Information Processing Systems (NeurIPS 2019)",
address = "United States",
url = "https://nips.cc/Conferences/2019",

}

RIS

TY - CHAP

T1 - Coresets for Archetypal Analysis

AU - Mair, Sebastian

AU - Brefeld, Ulf

N1 - Conference code: 33

PY - 2020

Y1 - 2020

N2 - Archetypal analysis represents instances as linear mixtures of prototypes (the archetypes) that lie on the boundary of the convex hull of the data. Archetypes are thus often better interpretable than factors computed by other matrix factorization techniques. However, the interpretability comes with high computational cost due to additional convexity-preserving constraints. In this paper, we propose efficient coresets for archetypal analysis. Theoretical guarantees are derived by showing that quantization errors of k-means upper bound archetypal analysis; the computation of a provable absolute-coreset can be performed in only two passes over the data. Empirically, we show that the coresets lead to improved performance on several data sets.

AB - Archetypal analysis represents instances as linear mixtures of prototypes (the archetypes) that lie on the boundary of the convex hull of the data. Archetypes are thus often better interpretable than factors computed by other matrix factorization techniques. However, the interpretability comes with high computational cost due to additional convexity-preserving constraints. In this paper, we propose efficient coresets for archetypal analysis. Theoretical guarantees are derived by showing that quantization errors of k-means upper bound archetypal analysis; the computation of a provable absolute-coreset can be performed in only two passes over the data. Empirically, we show that the coresets lead to improved performance on several data sets.

KW - Business informatics

UR - https://papers.nips.cc/paper/8945-coresets-for-archetypal-analysis

UR - https://proceedings.neurips.cc/paper/2019

UR - http://toc.proceedings.com/53719webtoc.pdf

UR - http://www.proceedings.com/53719.html

M3 - Article in conference proceedings

SN - 978-1-71380-793-3

VL - 10

T3 - Advances in neural information processing systems

SP - 7215

EP - 7223

BT - 32rd Conference on Neural Information Processing Systems (NeurIPS 2019)

A2 - Wallach, Hanna

A2 - Larochelle, Hugo

PB - Curran Associates

CY - Red Hook

T2 - 33rd Conference on Neural Information Processing Systems - NeurIPS 2019

Y2 - 8 December 2019 through 14 December 2019

ER -