Learning sparse penalties for change-point detection using max margin interval regression

Guillem Rigaill; Toby Dylan Hocking; Francis Bach; Jean Philippe Vert

Learning sparse penalties for change-point detection using max margin interval regression

Guillem Rigaill, Toby Dylan Hocking, Francis Bach, Jean Philippe Vert

Research output: Contribution to conference › Paper › peer-review

Abstract

In segmentation models, the number of change-points is typically chosen using a penalized cost function. In this work, we propose to learn the penalty and its constants in databases of signals with weak change-point annotations. We propose a convex relaxation for the resulting interval regression problem, and solve it using accelerated proximal gradient methods. We show that this method achieves state-of-the-art change-point detection in a database of annotated DNA copy number profiles from neuroblastoma tumors.

Original language	English (US)
Pages	1209-1217
Number of pages	9
State	Published - 2013
Externally published	Yes
Event	30th International Conference on Machine Learning, ICML 2013 - Atlanta, GA, United States Duration: Jun 16 2013 → Jun 21 2013

Conference

Conference	30th International Conference on Machine Learning, ICML 2013
Country/Territory	United States
City	Atlanta, GA
Period	6/16/13 → 6/21/13

ASJC Scopus subject areas

Human-Computer Interaction
Sociology and Political Science

Cite this

@conference{245b0a6382eb412e84e9fe42d410dccd,

title = "Learning sparse penalties for change-point detection using max margin interval regression",

abstract = "In segmentation models, the number of change-points is typically chosen using a penalized cost function. In this work, we propose to learn the penalty and its constants in databases of signals with weak change-point annotations. We propose a convex relaxation for the resulting interval regression problem, and solve it using accelerated proximal gradient methods. We show that this method achieves state-of-the-art change-point detection in a database of annotated DNA copy number profiles from neuroblastoma tumors.",

author = "Guillem Rigaill and Hocking, {Toby Dylan} and Francis Bach and Vert, {Jean Philippe}",

year = "2013",

language = "English (US)",

pages = "1209--1217",

note = "30th International Conference on Machine Learning, ICML 2013 ; Conference date: 16-06-2013 Through 21-06-2013",

}

TY - CONF

T1 - Learning sparse penalties for change-point detection using max margin interval regression

AU - Rigaill, Guillem

AU - Hocking, Toby Dylan

AU - Bach, Francis

AU - Vert, Jean Philippe

PY - 2013

Y1 - 2013

N2 - In segmentation models, the number of change-points is typically chosen using a penalized cost function. In this work, we propose to learn the penalty and its constants in databases of signals with weak change-point annotations. We propose a convex relaxation for the resulting interval regression problem, and solve it using accelerated proximal gradient methods. We show that this method achieves state-of-the-art change-point detection in a database of annotated DNA copy number profiles from neuroblastoma tumors.

AB - In segmentation models, the number of change-points is typically chosen using a penalized cost function. In this work, we propose to learn the penalty and its constants in databases of signals with weak change-point annotations. We propose a convex relaxation for the resulting interval regression problem, and solve it using accelerated proximal gradient methods. We show that this method achieves state-of-the-art change-point detection in a database of annotated DNA copy number profiles from neuroblastoma tumors.

UR - http://www.scopus.com/inward/record.url?scp=84897519123&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897519123&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:84897519123

SP - 1209

EP - 1217

T2 - 30th International Conference on Machine Learning, ICML 2013

Y2 - 16 June 2013 through 21 June 2013

ER -

Learning sparse penalties for change-point detection using max margin interval regression

Abstract

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this