Moderated and drifting linear dynamical systems

Jinyan Guan, Kyle Simek, Ernesto Brau, Clayton T Morrison, Emily A Butler, Jacobus J Barnard

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

1 Citation (Scopus)

Abstract

We consider linear dynamical systems, particularly coupled linear oscillators, where the parameters represent meaningful values in a domain theory, and thus learning what affects them contributes to explanation. Rather than allow perturbations of latent states, we assume that temporal variation beyond noise is explained by parameter drift, and variation across coupled systems is a function of moderating variables. This change in model structure reduces opportunities for efficient inference, and we propose sampling procedures to learn and fit the models. We test our approach on a real dataset of self-recalled emotional experience measurements of heterosexual couples engaged in a conversation about a potentially emotional topic, with body mass index (BMI) being considered as a moderator. We evaluate several models on their ability to predict future conversation dynamics (the last 20% of the data for each test couple), with shared parameters being learned using held out data. We validate the hypothesis that BMI affects the conversation dynamic in the experimentally chosen topic.

Original languageEnglish (US)
Title of host publication32nd International Conference on Machine Learning, ICML 2015
PublisherInternational Machine Learning Society (IMLS)
Pages2463-2472
Number of pages10
Volume3
ISBN (Print)9781510810587
StatePublished - 2015
Event32nd International Conference on Machine Learning, ICML 2015 - Lile, France
Duration: Jul 6 2015Jul 11 2015

Other

Other32nd International Conference on Machine Learning, ICML 2015
CountryFrance
CityLile
Period7/6/157/11/15

Fingerprint

Dynamical systems
Moderators
Model structures
Sampling

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Science Applications

Cite this

Guan, J., Simek, K., Brau, E., Morrison, C. T., Butler, E. A., & Barnard, J. J. (2015). Moderated and drifting linear dynamical systems. In 32nd International Conference on Machine Learning, ICML 2015 (Vol. 3, pp. 2463-2472). International Machine Learning Society (IMLS).

Moderated and drifting linear dynamical systems. / Guan, Jinyan; Simek, Kyle; Brau, Ernesto; Morrison, Clayton T; Butler, Emily A; Barnard, Jacobus J.

32nd International Conference on Machine Learning, ICML 2015. Vol. 3 International Machine Learning Society (IMLS), 2015. p. 2463-2472.

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

Guan, J, Simek, K, Brau, E, Morrison, CT, Butler, EA & Barnard, JJ 2015, Moderated and drifting linear dynamical systems. in 32nd International Conference on Machine Learning, ICML 2015. vol. 3, International Machine Learning Society (IMLS), pp. 2463-2472, 32nd International Conference on Machine Learning, ICML 2015, Lile, France, 7/6/15.
Guan J, Simek K, Brau E, Morrison CT, Butler EA, Barnard JJ. Moderated and drifting linear dynamical systems. In 32nd International Conference on Machine Learning, ICML 2015. Vol. 3. International Machine Learning Society (IMLS). 2015. p. 2463-2472
Guan, Jinyan ; Simek, Kyle ; Brau, Ernesto ; Morrison, Clayton T ; Butler, Emily A ; Barnard, Jacobus J. / Moderated and drifting linear dynamical systems. 32nd International Conference on Machine Learning, ICML 2015. Vol. 3 International Machine Learning Society (IMLS), 2015. pp. 2463-2472
@inproceedings{4f4a07e0e9f945439f5eb344a59f43f9,
title = "Moderated and drifting linear dynamical systems",
abstract = "We consider linear dynamical systems, particularly coupled linear oscillators, where the parameters represent meaningful values in a domain theory, and thus learning what affects them contributes to explanation. Rather than allow perturbations of latent states, we assume that temporal variation beyond noise is explained by parameter drift, and variation across coupled systems is a function of moderating variables. This change in model structure reduces opportunities for efficient inference, and we propose sampling procedures to learn and fit the models. We test our approach on a real dataset of self-recalled emotional experience measurements of heterosexual couples engaged in a conversation about a potentially emotional topic, with body mass index (BMI) being considered as a moderator. We evaluate several models on their ability to predict future conversation dynamics (the last 20{\%} of the data for each test couple), with shared parameters being learned using held out data. We validate the hypothesis that BMI affects the conversation dynamic in the experimentally chosen topic.",
author = "Jinyan Guan and Kyle Simek and Ernesto Brau and Morrison, {Clayton T} and Butler, {Emily A} and Barnard, {Jacobus J}",
year = "2015",
language = "English (US)",
isbn = "9781510810587",
volume = "3",
pages = "2463--2472",
booktitle = "32nd International Conference on Machine Learning, ICML 2015",
publisher = "International Machine Learning Society (IMLS)",

}

TY - GEN

T1 - Moderated and drifting linear dynamical systems

AU - Guan, Jinyan

AU - Simek, Kyle

AU - Brau, Ernesto

AU - Morrison, Clayton T

AU - Butler, Emily A

AU - Barnard, Jacobus J

PY - 2015

Y1 - 2015

N2 - We consider linear dynamical systems, particularly coupled linear oscillators, where the parameters represent meaningful values in a domain theory, and thus learning what affects them contributes to explanation. Rather than allow perturbations of latent states, we assume that temporal variation beyond noise is explained by parameter drift, and variation across coupled systems is a function of moderating variables. This change in model structure reduces opportunities for efficient inference, and we propose sampling procedures to learn and fit the models. We test our approach on a real dataset of self-recalled emotional experience measurements of heterosexual couples engaged in a conversation about a potentially emotional topic, with body mass index (BMI) being considered as a moderator. We evaluate several models on their ability to predict future conversation dynamics (the last 20% of the data for each test couple), with shared parameters being learned using held out data. We validate the hypothesis that BMI affects the conversation dynamic in the experimentally chosen topic.

AB - We consider linear dynamical systems, particularly coupled linear oscillators, where the parameters represent meaningful values in a domain theory, and thus learning what affects them contributes to explanation. Rather than allow perturbations of latent states, we assume that temporal variation beyond noise is explained by parameter drift, and variation across coupled systems is a function of moderating variables. This change in model structure reduces opportunities for efficient inference, and we propose sampling procedures to learn and fit the models. We test our approach on a real dataset of self-recalled emotional experience measurements of heterosexual couples engaged in a conversation about a potentially emotional topic, with body mass index (BMI) being considered as a moderator. We evaluate several models on their ability to predict future conversation dynamics (the last 20% of the data for each test couple), with shared parameters being learned using held out data. We validate the hypothesis that BMI affects the conversation dynamic in the experimentally chosen topic.

UR - http://www.scopus.com/inward/record.url?scp=84970021412&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84970021412&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781510810587

VL - 3

SP - 2463

EP - 2472

BT - 32nd International Conference on Machine Learning, ICML 2015

PB - International Machine Learning Society (IMLS)

ER -