Simpler knowledge-based support vector machines

Quoc V. Le; Alex J. Smola; Thomas Gärtner

doi:10.1145/1143844.1143910

Simpler knowledge-based support vector machines

Quoc V. Le^*, Alex J. Smola, Thomas Gärtner

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

12 Citations (Scopus)

Abstract

If appropriately used, prior knowledge can significantly improve the predictive accuracy of learning algorithms or reduce the amount of training data needed. In this paper we introduce a simple method to incorporate prior knowledge in support vector machines by modifying the hypothesis space rather than the optimization problem. The optimization problem is amenable to solution by the constrained concave convex procedure, which finds a local optimum. The paper discusses different kinds of prior knowledge and demonstrates the applicability of the approach in some characteristic experiments.

Original language	English
Title of host publication	ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006
Pages	521-528
Number of pages	8
DOIs	https://doi.org/10.1145/1143844.1143910
Publication status	Published - 2006
Event	23rd International Conference on Machine Learning, ICML 2006 - Pittsburgh, PA, United States Duration: 25 Jun 2006 → 29 Jun 2006

Publication series

Name	ACM International Conference Proceeding Series
Volume	148

Conference

Conference	23rd International Conference on Machine Learning, ICML 2006
Country/Territory	United States
City	Pittsburgh, PA
Period	25/06/06 → 29/06/06

Access to Document

10.1145/1143844.1143910

Cite this

@inproceedings{102228b444c74c31958afe0ddaacb894,

title = "Simpler knowledge-based support vector machines",

abstract = "If appropriately used, prior knowledge can significantly improve the predictive accuracy of learning algorithms or reduce the amount of training data needed. In this paper we introduce a simple method to incorporate prior knowledge in support vector machines by modifying the hypothesis space rather than the optimization problem. The optimization problem is amenable to solution by the constrained concave convex procedure, which finds a local optimum. The paper discusses different kinds of prior knowledge and demonstrates the applicability of the approach in some characteristic experiments.",

author = "Le, {Quoc V.} and Smola, {Alex J.} and Thomas G{\"a}rtner",

year = "2006",

doi = "10.1145/1143844.1143910",

language = "English",

isbn = "1595933832",

series = "ACM International Conference Proceeding Series",

pages = "521--528",

booktitle = "ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006",

note = "23rd International Conference on Machine Learning, ICML 2006 ; Conference date: 25-06-2006 Through 29-06-2006",

}

Le, QV, Smola, AJ & Gärtner, T 2006, Simpler knowledge-based support vector machines. in ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006. ACM International Conference Proceeding Series, vol. 148, pp. 521-528, 23rd International Conference on Machine Learning, ICML 2006, Pittsburgh, PA, United States, 25/06/06. https://doi.org/10.1145/1143844.1143910

Simpler knowledge-based support vector machines. / Le, Quoc V.; Smola, Alex J.; Gärtner, Thomas.
ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006. 2006. p. 521-528 (ACM International Conference Proceeding Series; Vol. 148).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Simpler knowledge-based support vector machines

AU - Le, Quoc V.

AU - Smola, Alex J.

AU - Gärtner, Thomas

PY - 2006

Y1 - 2006

N2 - If appropriately used, prior knowledge can significantly improve the predictive accuracy of learning algorithms or reduce the amount of training data needed. In this paper we introduce a simple method to incorporate prior knowledge in support vector machines by modifying the hypothesis space rather than the optimization problem. The optimization problem is amenable to solution by the constrained concave convex procedure, which finds a local optimum. The paper discusses different kinds of prior knowledge and demonstrates the applicability of the approach in some characteristic experiments.

AB - If appropriately used, prior knowledge can significantly improve the predictive accuracy of learning algorithms or reduce the amount of training data needed. In this paper we introduce a simple method to incorporate prior knowledge in support vector machines by modifying the hypothesis space rather than the optimization problem. The optimization problem is amenable to solution by the constrained concave convex procedure, which finds a local optimum. The paper discusses different kinds of prior knowledge and demonstrates the applicability of the approach in some characteristic experiments.

UR - http://www.scopus.com/inward/record.url?scp=34250789740&partnerID=8YFLogxK

U2 - 10.1145/1143844.1143910

DO - 10.1145/1143844.1143910

M3 - Conference contribution

SN - 1595933832

SN - 9781595933836

T3 - ACM International Conference Proceeding Series

SP - 521

EP - 528

BT - ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006

T2 - 23rd International Conference on Machine Learning, ICML 2006

Y2 - 25 June 2006 through 29 June 2006

ER -

Simpler knowledge-based support vector machines

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this