Home > Research > Publications & Outputs > Semi-Supervised Learning guided by the Generali...

Research

Electronic data

  • 2405.15294v2

    Final published version, 446 KB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

Keywords

View graph of relations

Semi-Supervised Learning guided by the Generalized Bayes Rule under Soft Revision

Research output: Working paper › Preprint

Published

Standard

Semi-Supervised Learning guided by the Generalized Bayes Rule under Soft Revision. / Dietrich, Stefan; Rodemann, Julian; Jansen, Christoph.
Arxiv, 2024.

Research output: Working paper › Preprint

Harvard

Dietrich, S, Rodemann, J & Jansen, C 2024 'Semi-Supervised Learning guided by the Generalized Bayes Rule under Soft Revision' Arxiv. <https://arxiv.org/abs/2405.15294v2>

APA

Dietrich, S., Rodemann, J., & Jansen, C. (2024). Semi-Supervised Learning guided by the Generalized Bayes Rule under Soft Revision. Arxiv. https://arxiv.org/abs/2405.15294v2

Vancouver

Dietrich S, Rodemann J, Jansen C. Semi-Supervised Learning guided by the Generalized Bayes Rule under Soft Revision. Arxiv. 2024 May 24.

Author

Dietrich, Stefan ; Rodemann, Julian ; Jansen, Christoph. / Semi-Supervised Learning guided by the Generalized Bayes Rule under Soft Revision. Arxiv, 2024.

Bibtex

@techreport{dd277da0ccc64e859a059e7d6efa0ec1,
title = "Semi-Supervised Learning guided by the Generalized Bayes Rule under Soft Revision",
abstract = " We provide a theoretical and computational investigation of the Gamma-Maximin method with soft revision, which was recently proposed as a robust criterion for pseudo-label selection (PLS) in semi-supervised learning. Opposed to traditional methods for PLS we use credal sets of priors ({"}generalized Bayes{"}) to represent the epistemic modeling uncertainty. These latter are then updated by the Gamma-Maximin method with soft revision. We eventually select pseudo-labeled data that are most likely in light of the least favorable distribution from the so updated credal set. We formalize the task of finding optimal pseudo-labeled data w.r.t. the Gamma-Maximin method with soft revision as an optimization problem. A concrete implementation for the class of logistic models then allows us to compare the predictive power of the method with competing approaches. It is observed that the Gamma-Maximin method with soft revision can achieve very promising results, especially when the proportion of labeled data is low. ",
keywords = "stat.ML, cs.AI, cs.LG, math.ST, stat.ME, stat.TH, 62C12 62C10, I.2.6; G.3",
author = "Stefan Dietrich and Julian Rodemann and Christoph Jansen",
note = "Accepted at the 11th International Conference on Soft Methods in Probability and Statistics (SMPS) 2024",
year = "2024",
month = may,
day = "24",
language = "English",
publisher = "Arxiv",
type = "WorkingPaper",
institution = "Arxiv",

}

RIS

TY - UNPB

T1 - Semi-Supervised Learning guided by the Generalized Bayes Rule under Soft Revision

AU - Dietrich, Stefan

AU - Rodemann, Julian

AU - Jansen, Christoph

N1 - Accepted at the 11th International Conference on Soft Methods in Probability and Statistics (SMPS) 2024

PY - 2024/5/24

Y1 - 2024/5/24

N2 - We provide a theoretical and computational investigation of the Gamma-Maximin method with soft revision, which was recently proposed as a robust criterion for pseudo-label selection (PLS) in semi-supervised learning. Opposed to traditional methods for PLS we use credal sets of priors ("generalized Bayes") to represent the epistemic modeling uncertainty. These latter are then updated by the Gamma-Maximin method with soft revision. We eventually select pseudo-labeled data that are most likely in light of the least favorable distribution from the so updated credal set. We formalize the task of finding optimal pseudo-labeled data w.r.t. the Gamma-Maximin method with soft revision as an optimization problem. A concrete implementation for the class of logistic models then allows us to compare the predictive power of the method with competing approaches. It is observed that the Gamma-Maximin method with soft revision can achieve very promising results, especially when the proportion of labeled data is low.

AB - We provide a theoretical and computational investigation of the Gamma-Maximin method with soft revision, which was recently proposed as a robust criterion for pseudo-label selection (PLS) in semi-supervised learning. Opposed to traditional methods for PLS we use credal sets of priors ("generalized Bayes") to represent the epistemic modeling uncertainty. These latter are then updated by the Gamma-Maximin method with soft revision. We eventually select pseudo-labeled data that are most likely in light of the least favorable distribution from the so updated credal set. We formalize the task of finding optimal pseudo-labeled data w.r.t. the Gamma-Maximin method with soft revision as an optimization problem. A concrete implementation for the class of logistic models then allows us to compare the predictive power of the method with competing approaches. It is observed that the Gamma-Maximin method with soft revision can achieve very promising results, especially when the proportion of labeled data is low.

KW - stat.ML

KW - cs.AI

KW - cs.LG

KW - math.ST

KW - stat.ME

KW - stat.TH

KW - 62C12 62C10

KW - I.2.6; G.3

M3 - Preprint

BT - Semi-Supervised Learning guided by the Generalized Bayes Rule under Soft Revision

PB - Arxiv

ER -