Hierarchical Fisher kernels for longitudinal data

Zhengdong Lu, Todd K. Leen, Jeffrey Kaye

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

1 Citation (Scopus)

Abstract

We develop new techniques for time series classification based on hierarchical Bayesian generative models (called mixed-effect models) and the Fisher kernel derived from them. A key advantage of the new formulation is that one can compute the Fisher information matrix despite varying sequence lengths and varying sampling intervals. This avoids the commonly-used ad hoc replacement of the Fisher information matrix with the identity which destroys the geometric invariance of the kernel. Our construction retains the geometric invariance, resulting in a kernel that is properly invariant under change of coordinates in the model parameter space. Experiments on detecting cognitive decline show that classifiers based on the proposed kernel out-perform those based on generative models and other feature extraction routines, and on Fisher kernels that use the identity in place of the Fisher information.

Original languageEnglish (US)
Title of host publicationAdvances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference
Pages1961-1968
Number of pages8
StatePublished - 2009
Event22nd Annual Conference on Neural Information Processing Systems, NIPS 2008 - Vancouver, BC, Canada
Duration: Dec 8 2008Dec 11 2008

Other

Other22nd Annual Conference on Neural Information Processing Systems, NIPS 2008
CountryCanada
CityVancouver, BC
Period12/8/0812/11/08

Fingerprint

Fisher information matrix
Invariance
Feature extraction
Time series
Classifiers
Sampling
Experiments

ASJC Scopus subject areas

  • Information Systems

Cite this

Lu, Z., Leen, T. K., & Kaye, J. (2009). Hierarchical Fisher kernels for longitudinal data. In Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference (pp. 1961-1968)

Hierarchical Fisher kernels for longitudinal data. / Lu, Zhengdong; Leen, Todd K.; Kaye, Jeffrey.

Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference. 2009. p. 1961-1968.

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

Lu, Z, Leen, TK & Kaye, J 2009, Hierarchical Fisher kernels for longitudinal data. in Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference. pp. 1961-1968, 22nd Annual Conference on Neural Information Processing Systems, NIPS 2008, Vancouver, BC, Canada, 12/8/08.
Lu Z, Leen TK, Kaye J. Hierarchical Fisher kernels for longitudinal data. In Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference. 2009. p. 1961-1968
Lu, Zhengdong ; Leen, Todd K. ; Kaye, Jeffrey. / Hierarchical Fisher kernels for longitudinal data. Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference. 2009. pp. 1961-1968
@inproceedings{e27b14f7ea1443a881870c76249e45fd,
title = "Hierarchical Fisher kernels for longitudinal data",
abstract = "We develop new techniques for time series classification based on hierarchical Bayesian generative models (called mixed-effect models) and the Fisher kernel derived from them. A key advantage of the new formulation is that one can compute the Fisher information matrix despite varying sequence lengths and varying sampling intervals. This avoids the commonly-used ad hoc replacement of the Fisher information matrix with the identity which destroys the geometric invariance of the kernel. Our construction retains the geometric invariance, resulting in a kernel that is properly invariant under change of coordinates in the model parameter space. Experiments on detecting cognitive decline show that classifiers based on the proposed kernel out-perform those based on generative models and other feature extraction routines, and on Fisher kernels that use the identity in place of the Fisher information.",
author = "Zhengdong Lu and Leen, {Todd K.} and Jeffrey Kaye",
year = "2009",
language = "English (US)",
isbn = "9781605609492",
pages = "1961--1968",
booktitle = "Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference",

}

TY - GEN

T1 - Hierarchical Fisher kernels for longitudinal data

AU - Lu, Zhengdong

AU - Leen, Todd K.

AU - Kaye, Jeffrey

PY - 2009

Y1 - 2009

N2 - We develop new techniques for time series classification based on hierarchical Bayesian generative models (called mixed-effect models) and the Fisher kernel derived from them. A key advantage of the new formulation is that one can compute the Fisher information matrix despite varying sequence lengths and varying sampling intervals. This avoids the commonly-used ad hoc replacement of the Fisher information matrix with the identity which destroys the geometric invariance of the kernel. Our construction retains the geometric invariance, resulting in a kernel that is properly invariant under change of coordinates in the model parameter space. Experiments on detecting cognitive decline show that classifiers based on the proposed kernel out-perform those based on generative models and other feature extraction routines, and on Fisher kernels that use the identity in place of the Fisher information.

AB - We develop new techniques for time series classification based on hierarchical Bayesian generative models (called mixed-effect models) and the Fisher kernel derived from them. A key advantage of the new formulation is that one can compute the Fisher information matrix despite varying sequence lengths and varying sampling intervals. This avoids the commonly-used ad hoc replacement of the Fisher information matrix with the identity which destroys the geometric invariance of the kernel. Our construction retains the geometric invariance, resulting in a kernel that is properly invariant under change of coordinates in the model parameter space. Experiments on detecting cognitive decline show that classifiers based on the proposed kernel out-perform those based on generative models and other feature extraction routines, and on Fisher kernels that use the identity in place of the Fisher information.

UR - http://www.scopus.com/inward/record.url?scp=84858767817&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84858767817&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781605609492

SP - 1961

EP - 1968

BT - Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference

ER -