Label Propagation for Learning with Label Proportions

Rafael Poyiadzi; Raul Santos-Rodriguez; Niall Twomey

doi:10.1109/MLSP.2018.8517083

Label Propagation for Learning with Label Proportions

Rafael Poyiadzi, Raul Santos-Rodriguez, Niall Twomey

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

267 Downloads (Pure)

Abstract

Learning with Label Proportions (LLP) is the problem of recovering the underlying true labels given a dataset when the data is presented in the form of bags. This paradigm is particularly suitable in contexts where providing individual labels is expensive and label aggregates are more easily obtained. In the healthcare domain, it is a burden for a patient to keep a detailed diary of their daily routines, but often they will be amenable to provide higher level summaries of daily behavior. We present a novel and efficient graph-based algorithm that encourages local smoothness and exploits the global structure of the data, while preserving the 'mass' of each bag.

Original language	English
Title of host publication	2018 IEEE 28th International Workshop on Machine Learning for Signal Processing (MLSP 2018)
Subtitle of host publication	Proceedings of a meeting held 17-20 September 2018, Aalborg, Denmark
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	264-270
Number of pages	6
ISBN (Electronic)	9781538654774
ISBN (Print)	9781538654781
DOIs	https://doi.org/10.1109/MLSP.2018.8517083
Publication status	Published - Nov 2018

Publication series

Name
ISSN (Print)	1551-2541

Structured keywords

Digital Health

Access to Document

10.1109/MLSP.2018.8517083

Full-text PDF (final published version)
This is the final published version of the article (version of record). It first appeared online via IEEE at https://ieeexplore.ieee.org/document/8517083 . Please refer to any applicable terms of use of the publisher.
Final published version, 376 KBLicence: CC BY-NC-SA

https://arxiv.org/abs/1810.10328

Fingerprint Dive into the research topics of 'Label Propagation for Learning with Label Proportions'. Together they form a unique fingerprint.

Projects

2 Finished

Momentum Award - The use of home sensors in the early diagnosis and monitoring of Dementia (CUBOId)

Ben-Shlomo, Y.

1/04/17 → 31/03/20

Project: Research, Parent

SPHERE (EPSRC IRC)

Craddock, I. J., Coyle, D. T., Flach, P. A., Kaleshi, D., Mirmehdi, M., Piechocki, R. J., Stark, B. H., Ascione, R., Ashburn, A. M., Burnett, M. E., Aldamen, D., Gooberman-Hill, R. J. S., Harwin, W. S., Hilton, G., Holderbaum, W., Holley, A. P., Manchester, V. A., Meller, B. J., Stack, E. & Gilchrist, I. D.

1/10/13 → 30/09/18

Project: Research, Parent

Cite this

Poyiadzi, R., Santos-Rodriguez, R., & Twomey, N. (2018). Label Propagation for Learning with Label Proportions. In 2018 IEEE 28th International Workshop on Machine Learning for Signal Processing (MLSP 2018): Proceedings of a meeting held 17-20 September 2018, Aalborg, Denmark (pp. 264-270). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/MLSP.2018.8517083

@inproceedings{ed8fc638d7a246a8b2f26ed1358159f1,

title = "Label Propagation for Learning with Label Proportions",

abstract = "Learning with Label Proportions (LLP) is the problem of recovering the underlying true labels given a dataset when the data is presented in the form of bags. This paradigm is particularly suitable in contexts where providing individual labels is expensive and label aggregates are more easily obtained. In the healthcare domain, it is a burden for a patient to keep a detailed diary of their daily routines, but often they will be amenable to provide higher level summaries of daily behavior. We present a novel and efficient graph-based algorithm that encourages local smoothness and exploits the global structure of the data, while preserving the 'mass' of each bag.",

author = "Rafael Poyiadzi and Raul Santos-Rodriguez and Niall Twomey",

year = "2018",

month = nov,

doi = "10.1109/MLSP.2018.8517083",

language = "English",

isbn = "9781538654781",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

pages = "264--270",

booktitle = "2018 IEEE 28th International Workshop on Machine Learning for Signal Processing (MLSP 2018)",

address = "United States",

}

Poyiadzi, R, Santos-Rodriguez, R & Twomey, N 2018, Label Propagation for Learning with Label Proportions. in 2018 IEEE 28th International Workshop on Machine Learning for Signal Processing (MLSP 2018): Proceedings of a meeting held 17-20 September 2018, Aalborg, Denmark. Institute of Electrical and Electronics Engineers (IEEE), pp. 264-270. https://doi.org/10.1109/MLSP.2018.8517083

Label Propagation for Learning with Label Proportions. / Poyiadzi, Rafael; Santos-Rodriguez, Raul; Twomey, Niall.

2018 IEEE 28th International Workshop on Machine Learning for Signal Processing (MLSP 2018): Proceedings of a meeting held 17-20 September 2018, Aalborg, Denmark. Institute of Electrical and Electronics Engineers (IEEE), 2018. p. 264-270.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Label Propagation for Learning with Label Proportions

AU - Poyiadzi, Rafael

AU - Santos-Rodriguez, Raul

AU - Twomey, Niall

PY - 2018/11

Y1 - 2018/11

N2 - Learning with Label Proportions (LLP) is the problem of recovering the underlying true labels given a dataset when the data is presented in the form of bags. This paradigm is particularly suitable in contexts where providing individual labels is expensive and label aggregates are more easily obtained. In the healthcare domain, it is a burden for a patient to keep a detailed diary of their daily routines, but often they will be amenable to provide higher level summaries of daily behavior. We present a novel and efficient graph-based algorithm that encourages local smoothness and exploits the global structure of the data, while preserving the 'mass' of each bag.

AB - Learning with Label Proportions (LLP) is the problem of recovering the underlying true labels given a dataset when the data is presented in the form of bags. This paradigm is particularly suitable in contexts where providing individual labels is expensive and label aggregates are more easily obtained. In the healthcare domain, it is a burden for a patient to keep a detailed diary of their daily routines, but often they will be amenable to provide higher level summaries of daily behavior. We present a novel and efficient graph-based algorithm that encourages local smoothness and exploits the global structure of the data, while preserving the 'mass' of each bag.

UR - https://arxiv.org/abs/1810.10328

U2 - 10.1109/MLSP.2018.8517083

DO - 10.1109/MLSP.2018.8517083

M3 - Conference Contribution (Conference Proceeding)

SN - 9781538654781

SP - 264

EP - 270

BT - 2018 IEEE 28th International Workshop on Machine Learning for Signal Processing (MLSP 2018)

PB - Institute of Electrical and Electronics Engineers (IEEE)

ER -

Poyiadzi R, Santos-Rodriguez R, Twomey N. Label Propagation for Learning with Label Proportions. In 2018 IEEE 28th International Workshop on Machine Learning for Signal Processing (MLSP 2018): Proceedings of a meeting held 17-20 September 2018, Aalborg, Denmark. Institute of Electrical and Electronics Engineers (IEEE). 2018. p. 264-270 https://doi.org/10.1109/MLSP.2018.8517083