Weighted automata kernels - General framework and algorithms

Corinna Cortes, Patrick Haffner, Mehryar Mohri

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

Abstract

Kernel methods have found in recent years wide use in statistical learning techniques due to their good performance and their computational efficiency in high-dimensional feature space. However, text or speech data cannot always be represented by the fixed-length vectors that the traditional kernels handle. We recently introduced a general kernel framework based on weighted transducers, rational kernels, to extend kernel methods to the analysis of variable-length sequences and weighted automata [5] and described their application to spoken-dialog applications. We presented a constructive algorithm for ensuring that rational kernels are positive definite symmetric, a property which guarantees the convergence of discriminant classification algorithms such as Support Vector Machines, and showed that many string kernels previously introduced in the computational biology literature are special instances of such positive definite symmetric rational kernels [4]. This paper reviews the essential results given in [5, 3, 4] and presents them in the form of a short tutorial.

Original languageEnglish (US)
Title of host publicationEUROSPEECH 2003 - 8th European Conference on Speech Communication and Technology
PublisherInternational Speech Communication Association
Pages989-992
Number of pages4
StatePublished - 2003
Event8th European Conference on Speech Communication and Technology, EUROSPEECH 2003 - Geneva, Switzerland
Duration: Sep 1 2003Sep 4 2003

Other

Other8th European Conference on Speech Communication and Technology, EUROSPEECH 2003
CountrySwitzerland
CityGeneva
Period9/1/039/4/03

Fingerprint

Computational efficiency
Support vector machines
biology
Transducers
guarantee
dialogue
efficiency
learning
performance
literature

ASJC Scopus subject areas

  • Computer Science Applications
  • Software
  • Linguistics and Language
  • Communication

Cite this

Cortes, C., Haffner, P., & Mohri, M. (2003). Weighted automata kernels - General framework and algorithms. In EUROSPEECH 2003 - 8th European Conference on Speech Communication and Technology (pp. 989-992). International Speech Communication Association.

Weighted automata kernels - General framework and algorithms. / Cortes, Corinna; Haffner, Patrick; Mohri, Mehryar.

EUROSPEECH 2003 - 8th European Conference on Speech Communication and Technology. International Speech Communication Association, 2003. p. 989-992.

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

Cortes, C, Haffner, P & Mohri, M 2003, Weighted automata kernels - General framework and algorithms. in EUROSPEECH 2003 - 8th European Conference on Speech Communication and Technology. International Speech Communication Association, pp. 989-992, 8th European Conference on Speech Communication and Technology, EUROSPEECH 2003, Geneva, Switzerland, 9/1/03.
Cortes C, Haffner P, Mohri M. Weighted automata kernels - General framework and algorithms. In EUROSPEECH 2003 - 8th European Conference on Speech Communication and Technology. International Speech Communication Association. 2003. p. 989-992
Cortes, Corinna ; Haffner, Patrick ; Mohri, Mehryar. / Weighted automata kernels - General framework and algorithms. EUROSPEECH 2003 - 8th European Conference on Speech Communication and Technology. International Speech Communication Association, 2003. pp. 989-992
@inproceedings{85094f12c6f9466491e42ee3b1110ca4,
title = "Weighted automata kernels - General framework and algorithms",
abstract = "Kernel methods have found in recent years wide use in statistical learning techniques due to their good performance and their computational efficiency in high-dimensional feature space. However, text or speech data cannot always be represented by the fixed-length vectors that the traditional kernels handle. We recently introduced a general kernel framework based on weighted transducers, rational kernels, to extend kernel methods to the analysis of variable-length sequences and weighted automata [5] and described their application to spoken-dialog applications. We presented a constructive algorithm for ensuring that rational kernels are positive definite symmetric, a property which guarantees the convergence of discriminant classification algorithms such as Support Vector Machines, and showed that many string kernels previously introduced in the computational biology literature are special instances of such positive definite symmetric rational kernels [4]. This paper reviews the essential results given in [5, 3, 4] and presents them in the form of a short tutorial.",
author = "Corinna Cortes and Patrick Haffner and Mehryar Mohri",
year = "2003",
language = "English (US)",
pages = "989--992",
booktitle = "EUROSPEECH 2003 - 8th European Conference on Speech Communication and Technology",
publisher = "International Speech Communication Association",

}

TY - GEN

T1 - Weighted automata kernels - General framework and algorithms

AU - Cortes, Corinna

AU - Haffner, Patrick

AU - Mohri, Mehryar

PY - 2003

Y1 - 2003

N2 - Kernel methods have found in recent years wide use in statistical learning techniques due to their good performance and their computational efficiency in high-dimensional feature space. However, text or speech data cannot always be represented by the fixed-length vectors that the traditional kernels handle. We recently introduced a general kernel framework based on weighted transducers, rational kernels, to extend kernel methods to the analysis of variable-length sequences and weighted automata [5] and described their application to spoken-dialog applications. We presented a constructive algorithm for ensuring that rational kernels are positive definite symmetric, a property which guarantees the convergence of discriminant classification algorithms such as Support Vector Machines, and showed that many string kernels previously introduced in the computational biology literature are special instances of such positive definite symmetric rational kernels [4]. This paper reviews the essential results given in [5, 3, 4] and presents them in the form of a short tutorial.

AB - Kernel methods have found in recent years wide use in statistical learning techniques due to their good performance and their computational efficiency in high-dimensional feature space. However, text or speech data cannot always be represented by the fixed-length vectors that the traditional kernels handle. We recently introduced a general kernel framework based on weighted transducers, rational kernels, to extend kernel methods to the analysis of variable-length sequences and weighted automata [5] and described their application to spoken-dialog applications. We presented a constructive algorithm for ensuring that rational kernels are positive definite symmetric, a property which guarantees the convergence of discriminant classification algorithms such as Support Vector Machines, and showed that many string kernels previously introduced in the computational biology literature are special instances of such positive definite symmetric rational kernels [4]. This paper reviews the essential results given in [5, 3, 4] and presents them in the form of a short tutorial.

UR - http://www.scopus.com/inward/record.url?scp=85009198583&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009198583&partnerID=8YFLogxK

M3 - Conference contribution

SP - 989

EP - 992

BT - EUROSPEECH 2003 - 8th European Conference on Speech Communication and Technology

PB - International Speech Communication Association

ER -