Gaussian Graphical Model Selection from Size Constrained Measurements

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we introduce the problem of learning graphical models from size constrained measurements. This is inspired by a wide range of problems where one is unable to measure all the variables involved simultaneously. We propose notions of data requirement for this setting and then begin by considering an extreme case where one is allowed to only measure pairs of variables. For this setting we propose a simple algorithm and provide guarantees on its behavior. We then generalize to the case where one is allowed to measure up to r variables simultaneously, and draw connections to the field of combinatorial designs. Finally, we propose an interactive version of the proposed algorithm that is guaranteed to have significantly better data requirement on a wide range of realistic settings.

Original languageEnglish (US)
Title of host publication2019 IEEE International Symposium on Information Theory, ISIT 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1302-1306
Number of pages5
ISBN (Electronic)9781538692912
DOIs
StatePublished - Jul 2019
Event2019 IEEE International Symposium on Information Theory, ISIT 2019 - Paris, France
Duration: Jul 7 2019Jul 12 2019

Publication series

NameIEEE International Symposium on Information Theory - Proceedings
Volume2019-July
ISSN (Print)2157-8095

Conference

Conference2019 IEEE International Symposium on Information Theory, ISIT 2019
CountryFrance
CityParis
Period7/7/197/12/19

Fingerprint

Gaussian Model
Graphical Models
Model Selection
Combinatorial Design
Requirements
Range of data
Extremes
Generalise

Keywords

  • active learning
  • combinatorial designs
  • Gaussian graphical models
  • sample complexity

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Information Systems
  • Modeling and Simulation
  • Applied Mathematics

Cite this

Dasarathy, G. (2019). Gaussian Graphical Model Selection from Size Constrained Measurements. In 2019 IEEE International Symposium on Information Theory, ISIT 2019 - Proceedings (pp. 1302-1306). [8849299] (IEEE International Symposium on Information Theory - Proceedings; Vol. 2019-July). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISIT.2019.8849299

Gaussian Graphical Model Selection from Size Constrained Measurements. / Dasarathy, Gautam.

2019 IEEE International Symposium on Information Theory, ISIT 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 1302-1306 8849299 (IEEE International Symposium on Information Theory - Proceedings; Vol. 2019-July).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Dasarathy, G 2019, Gaussian Graphical Model Selection from Size Constrained Measurements. in 2019 IEEE International Symposium on Information Theory, ISIT 2019 - Proceedings., 8849299, IEEE International Symposium on Information Theory - Proceedings, vol. 2019-July, Institute of Electrical and Electronics Engineers Inc., pp. 1302-1306, 2019 IEEE International Symposium on Information Theory, ISIT 2019, Paris, France, 7/7/19. https://doi.org/10.1109/ISIT.2019.8849299
Dasarathy G. Gaussian Graphical Model Selection from Size Constrained Measurements. In 2019 IEEE International Symposium on Information Theory, ISIT 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 1302-1306. 8849299. (IEEE International Symposium on Information Theory - Proceedings). https://doi.org/10.1109/ISIT.2019.8849299
Dasarathy, Gautam. / Gaussian Graphical Model Selection from Size Constrained Measurements. 2019 IEEE International Symposium on Information Theory, ISIT 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 1302-1306 (IEEE International Symposium on Information Theory - Proceedings).
@inproceedings{84ffef30bf7e4489bd8ef1509ed8dd35,
title = "Gaussian Graphical Model Selection from Size Constrained Measurements",
abstract = "In this paper, we introduce the problem of learning graphical models from size constrained measurements. This is inspired by a wide range of problems where one is unable to measure all the variables involved simultaneously. We propose notions of data requirement for this setting and then begin by considering an extreme case where one is allowed to only measure pairs of variables. For this setting we propose a simple algorithm and provide guarantees on its behavior. We then generalize to the case where one is allowed to measure up to r variables simultaneously, and draw connections to the field of combinatorial designs. Finally, we propose an interactive version of the proposed algorithm that is guaranteed to have significantly better data requirement on a wide range of realistic settings.",
keywords = "active learning, combinatorial designs, Gaussian graphical models, sample complexity",
author = "Gautam Dasarathy",
year = "2019",
month = "7",
doi = "10.1109/ISIT.2019.8849299",
language = "English (US)",
series = "IEEE International Symposium on Information Theory - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "1302--1306",
booktitle = "2019 IEEE International Symposium on Information Theory, ISIT 2019 - Proceedings",

}

TY - GEN

T1 - Gaussian Graphical Model Selection from Size Constrained Measurements

AU - Dasarathy, Gautam

PY - 2019/7

Y1 - 2019/7

N2 - In this paper, we introduce the problem of learning graphical models from size constrained measurements. This is inspired by a wide range of problems where one is unable to measure all the variables involved simultaneously. We propose notions of data requirement for this setting and then begin by considering an extreme case where one is allowed to only measure pairs of variables. For this setting we propose a simple algorithm and provide guarantees on its behavior. We then generalize to the case where one is allowed to measure up to r variables simultaneously, and draw connections to the field of combinatorial designs. Finally, we propose an interactive version of the proposed algorithm that is guaranteed to have significantly better data requirement on a wide range of realistic settings.

AB - In this paper, we introduce the problem of learning graphical models from size constrained measurements. This is inspired by a wide range of problems where one is unable to measure all the variables involved simultaneously. We propose notions of data requirement for this setting and then begin by considering an extreme case where one is allowed to only measure pairs of variables. For this setting we propose a simple algorithm and provide guarantees on its behavior. We then generalize to the case where one is allowed to measure up to r variables simultaneously, and draw connections to the field of combinatorial designs. Finally, we propose an interactive version of the proposed algorithm that is guaranteed to have significantly better data requirement on a wide range of realistic settings.

KW - active learning

KW - combinatorial designs

KW - Gaussian graphical models

KW - sample complexity

UR - http://www.scopus.com/inward/record.url?scp=85073155023&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073155023&partnerID=8YFLogxK

U2 - 10.1109/ISIT.2019.8849299

DO - 10.1109/ISIT.2019.8849299

M3 - Conference contribution

AN - SCOPUS:85073155023

T3 - IEEE International Symposium on Information Theory - Proceedings

SP - 1302

EP - 1306

BT - 2019 IEEE International Symposium on Information Theory, ISIT 2019 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -