Generative multiple-instance learning models for quantitative electromyography

Tameem Adel, Ruth Urner, Benn Smith, Daniel Stashuk, Daniel J. Lizotte

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

7 Citations (Scopus)

Abstract

We present a comprehensive study of the use of generative modeling approaches for Multiple-Instance Learning (MIL) problems. In MIL a learner receives training instances grouped together into bags with labels for the bags only (which might not be correct for the comprised instances). Our work was motivated by the task of facilitating the diagnosis of neuromuscular disorders using sets of motor unit potential trains (MUPTs) detected within a muscle which can be cast as a MIL problem. Our approach leads to a state-of-the-art solution to the problem of muscle classification. By introducing and analyzing generative models for MIL in a general framework and examining a variety of model structures and components, our work also serves as a methodological guide to modelling MIL tasks. We evaluate our proposed methods both on MUPT datasets and on the MUSK1 dataset, one of the most widely used benchmarks for MIL.

Original languageEnglish (US)
Title of host publicationUncertainty in Artificial Intelligence - Proceedings of the 29th Conference, UAI 2013
Pages2-11
Number of pages10
StatePublished - 2013
Event29th Conference on Uncertainty in Artificial Intelligence, UAI 2013 - Bellevue, WA, United States
Duration: Jul 11 2013Jul 15 2013

Other

Other29th Conference on Uncertainty in Artificial Intelligence, UAI 2013
CountryUnited States
CityBellevue, WA
Period7/11/137/15/13

Fingerprint

Electromyography
Muscle
Model structures
Labels

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Adel, T., Urner, R., Smith, B., Stashuk, D., & Lizotte, D. J. (2013). Generative multiple-instance learning models for quantitative electromyography. In Uncertainty in Artificial Intelligence - Proceedings of the 29th Conference, UAI 2013 (pp. 2-11)

Generative multiple-instance learning models for quantitative electromyography. / Adel, Tameem; Urner, Ruth; Smith, Benn; Stashuk, Daniel; Lizotte, Daniel J.

Uncertainty in Artificial Intelligence - Proceedings of the 29th Conference, UAI 2013. 2013. p. 2-11.

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

Adel, T, Urner, R, Smith, B, Stashuk, D & Lizotte, DJ 2013, Generative multiple-instance learning models for quantitative electromyography. in Uncertainty in Artificial Intelligence - Proceedings of the 29th Conference, UAI 2013. pp. 2-11, 29th Conference on Uncertainty in Artificial Intelligence, UAI 2013, Bellevue, WA, United States, 7/11/13.
Adel T, Urner R, Smith B, Stashuk D, Lizotte DJ. Generative multiple-instance learning models for quantitative electromyography. In Uncertainty in Artificial Intelligence - Proceedings of the 29th Conference, UAI 2013. 2013. p. 2-11
Adel, Tameem ; Urner, Ruth ; Smith, Benn ; Stashuk, Daniel ; Lizotte, Daniel J. / Generative multiple-instance learning models for quantitative electromyography. Uncertainty in Artificial Intelligence - Proceedings of the 29th Conference, UAI 2013. 2013. pp. 2-11
@inproceedings{b29578b8104b47bfa8f72fea5bcd624b,
title = "Generative multiple-instance learning models for quantitative electromyography",
abstract = "We present a comprehensive study of the use of generative modeling approaches for Multiple-Instance Learning (MIL) problems. In MIL a learner receives training instances grouped together into bags with labels for the bags only (which might not be correct for the comprised instances). Our work was motivated by the task of facilitating the diagnosis of neuromuscular disorders using sets of motor unit potential trains (MUPTs) detected within a muscle which can be cast as a MIL problem. Our approach leads to a state-of-the-art solution to the problem of muscle classification. By introducing and analyzing generative models for MIL in a general framework and examining a variety of model structures and components, our work also serves as a methodological guide to modelling MIL tasks. We evaluate our proposed methods both on MUPT datasets and on the MUSK1 dataset, one of the most widely used benchmarks for MIL.",
author = "Tameem Adel and Ruth Urner and Benn Smith and Daniel Stashuk and Lizotte, {Daniel J.}",
year = "2013",
language = "English (US)",
pages = "2--11",
booktitle = "Uncertainty in Artificial Intelligence - Proceedings of the 29th Conference, UAI 2013",

}

TY - GEN

T1 - Generative multiple-instance learning models for quantitative electromyography

AU - Adel, Tameem

AU - Urner, Ruth

AU - Smith, Benn

AU - Stashuk, Daniel

AU - Lizotte, Daniel J.

PY - 2013

Y1 - 2013

N2 - We present a comprehensive study of the use of generative modeling approaches for Multiple-Instance Learning (MIL) problems. In MIL a learner receives training instances grouped together into bags with labels for the bags only (which might not be correct for the comprised instances). Our work was motivated by the task of facilitating the diagnosis of neuromuscular disorders using sets of motor unit potential trains (MUPTs) detected within a muscle which can be cast as a MIL problem. Our approach leads to a state-of-the-art solution to the problem of muscle classification. By introducing and analyzing generative models for MIL in a general framework and examining a variety of model structures and components, our work also serves as a methodological guide to modelling MIL tasks. We evaluate our proposed methods both on MUPT datasets and on the MUSK1 dataset, one of the most widely used benchmarks for MIL.

AB - We present a comprehensive study of the use of generative modeling approaches for Multiple-Instance Learning (MIL) problems. In MIL a learner receives training instances grouped together into bags with labels for the bags only (which might not be correct for the comprised instances). Our work was motivated by the task of facilitating the diagnosis of neuromuscular disorders using sets of motor unit potential trains (MUPTs) detected within a muscle which can be cast as a MIL problem. Our approach leads to a state-of-the-art solution to the problem of muscle classification. By introducing and analyzing generative models for MIL in a general framework and examining a variety of model structures and components, our work also serves as a methodological guide to modelling MIL tasks. We evaluate our proposed methods both on MUPT datasets and on the MUSK1 dataset, one of the most widely used benchmarks for MIL.

UR - http://www.scopus.com/inward/record.url?scp=84888192379&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84888192379&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84888192379

SP - 2

EP - 11

BT - Uncertainty in Artificial Intelligence - Proceedings of the 29th Conference, UAI 2013

ER -