Semi-supervised learning for structured output variables

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Standard

Semi-supervised learning for structured output variables. / Brefeld, Ulf; Scheffer, Tobias.
Proceedings of the 23rd international conference on Machine learning. Hrsg. / William Cohen; Andrew Moore. Association for Computing Machinery, Inc, 2006. S. 145-152.

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Harvard

Brefeld, U & Scheffer, T 2006, Semi-supervised learning for structured output variables. in W Cohen & A Moore (Hrsg.), Proceedings of the 23rd international conference on Machine learning. Association for Computing Machinery, Inc, S. 145-152, International Conference on Machine Learning - ICML 2006, Pittsburgh, USA / Vereinigte Staaten, 25.06.06. https://doi.org/10.1145/1143844.1143863

APA

Brefeld, U., & Scheffer, T. (2006). Semi-supervised learning for structured output variables. In W. Cohen, & A. Moore (Hrsg.), Proceedings of the 23rd international conference on Machine learning (S. 145-152). Association for Computing Machinery, Inc. https://doi.org/10.1145/1143844.1143863

Vancouver

Brefeld U, Scheffer T. Semi-supervised learning for structured output variables. in Cohen W, Moore A, Hrsg., Proceedings of the 23rd international conference on Machine learning. Association for Computing Machinery, Inc. 2006. S. 145-152 doi: 10.1145/1143844.1143863

Bibtex

@inbook{f5efe23e24b4435d9f0359def033ae8a,
title = "Semi-supervised learning for structured output variables",
abstract = "The problem of learning a mapping between input and structured, interdependent output variables covers sequential, spatial, and relational learning as well as predicting recursive structures. Joint feature representations of the input and output variables have paved the way to leveraging discriminative learners such as SVMs to this class of problems. We address the problem of semi-supervised learning in joint input output spaces. The cotraining approach is based on the principle of maximizing the consensus among multiple independent hypotheses; we develop this principle into a semi-supervised support vector learning algorithm for joint input out-put spaces and arbitrary loss functions. Experiments investigate the benefit of semi-supervised structured models in terms of accuracy and F1 score.",
keywords = "Informatics, Business informatics",
author = "Ulf Brefeld and Tobias Scheffer",
year = "2006",
month = jan,
day = "1",
doi = "10.1145/1143844.1143863",
language = "English",
isbn = "978-1-59593-383-6",
pages = "145--152",
editor = "William Cohen and Andrew Moore",
booktitle = "Proceedings of the 23rd international conference on Machine learning",
publisher = "Association for Computing Machinery, Inc",
address = "United States",
note = "ICML '06 ; Conference date: 25-06-2006 Through 29-06-2006",

}

RIS

TY - CHAP

T1 - Semi-supervised learning for structured output variables

AU - Brefeld, Ulf

AU - Scheffer, Tobias

N1 - Conference code: 23

PY - 2006/1/1

Y1 - 2006/1/1

N2 - The problem of learning a mapping between input and structured, interdependent output variables covers sequential, spatial, and relational learning as well as predicting recursive structures. Joint feature representations of the input and output variables have paved the way to leveraging discriminative learners such as SVMs to this class of problems. We address the problem of semi-supervised learning in joint input output spaces. The cotraining approach is based on the principle of maximizing the consensus among multiple independent hypotheses; we develop this principle into a semi-supervised support vector learning algorithm for joint input out-put spaces and arbitrary loss functions. Experiments investigate the benefit of semi-supervised structured models in terms of accuracy and F1 score.

AB - The problem of learning a mapping between input and structured, interdependent output variables covers sequential, spatial, and relational learning as well as predicting recursive structures. Joint feature representations of the input and output variables have paved the way to leveraging discriminative learners such as SVMs to this class of problems. We address the problem of semi-supervised learning in joint input output spaces. The cotraining approach is based on the principle of maximizing the consensus among multiple independent hypotheses; we develop this principle into a semi-supervised support vector learning algorithm for joint input out-put spaces and arbitrary loss functions. Experiments investigate the benefit of semi-supervised structured models in terms of accuracy and F1 score.

KW - Informatics

KW - Business informatics

UR - http://www.scopus.com/inward/record.url?scp=34250753883&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/ddd5325c-60a2-384e-b277-453f35a64a0e/

U2 - 10.1145/1143844.1143863

DO - 10.1145/1143844.1143863

M3 - Article in conference proceedings

AN - SCOPUS:34250753883

SN - 978-1-59593-383-6

SP - 145

EP - 152

BT - Proceedings of the 23rd international conference on Machine learning

A2 - Cohen, William

A2 - Moore, Andrew

PB - Association for Computing Machinery, Inc

T2 - ICML '06

Y2 - 25 June 2006 through 29 June 2006

ER -

DOI