Time series filtering, smoothing and learning using the kernel Kalman filter

Liva Ralaivola; Florence D'Alché-Buc

doi:10.1109/IJCNN.2005.1556088

Time series filtering, smoothing and learning using the kernel Kalman filter

Liva Ralaivola
, Florence D'Alché-Buc

Aix Marseille Université
Université d'Evry Val d'Essonne

Research output: Contribution to conference › Paper › peer-review

Abstract

In this paper, we propose a new model, the Kernel Kalman Filter, to perform various nonlinear time series processing. This model is based on the use of Mercer kernel functions in the framework of the Kalman Filter or Linear Dynamical Systems. Thanks to the kernel trick, all the equations involved in our model to perform filtering, smoothing and learning tasks, only require matrix algebra calculus whilst providing the ability to model complex time series. In particular, it is possible to learn dynamics from some nonlinear noisy time series implementing an exact Expectation-Maximization procedure.

Original language	English
Pages	1454-1459
Number of pages	6
DOIs	https://doi.org/10.1109/IJCNN.2005.1556088
Publication status	Published - 1 Jan 2005
Externally published	Yes
Event	International Joint Conference on Neural Networks, IJCNN 2005 - Montreal, QC, Canada Duration: 31 Jul 2005 → 4 Aug 2005

Conference

Conference	International Joint Conference on Neural Networks, IJCNN 2005
Country/Territory	Canada
City	Montreal, QC
Period	31/07/05 → 4/08/05

Access to Document

10.1109/IJCNN.2005.1556088

Cite this

@conference{6884e201a02f4674b030489b0689701c,

title = "Time series filtering, smoothing and learning using the kernel Kalman filter",

abstract = "In this paper, we propose a new model, the Kernel Kalman Filter, to perform various nonlinear time series processing. This model is based on the use of Mercer kernel functions in the framework of the Kalman Filter or Linear Dynamical Systems. Thanks to the kernel trick, all the equations involved in our model to perform filtering, smoothing and learning tasks, only require matrix algebra calculus whilst providing the ability to model complex time series. In particular, it is possible to learn dynamics from some nonlinear noisy time series implementing an exact Expectation-Maximization procedure.",

author = "Liva Ralaivola and Florence D'Alch{\'e}-Buc",

year = "2005",

month = jan,

day = "1",

doi = "10.1109/IJCNN.2005.1556088",

language = "English",

pages = "1454--1459",

note = "International Joint Conference on Neural Networks, IJCNN 2005 ; Conference date: 31-07-2005 Through 04-08-2005",

}

TY - CONF

T1 - Time series filtering, smoothing and learning using the kernel Kalman filter

AU - Ralaivola, Liva

AU - D'Alché-Buc, Florence

PY - 2005/1/1

Y1 - 2005/1/1

N2 - In this paper, we propose a new model, the Kernel Kalman Filter, to perform various nonlinear time series processing. This model is based on the use of Mercer kernel functions in the framework of the Kalman Filter or Linear Dynamical Systems. Thanks to the kernel trick, all the equations involved in our model to perform filtering, smoothing and learning tasks, only require matrix algebra calculus whilst providing the ability to model complex time series. In particular, it is possible to learn dynamics from some nonlinear noisy time series implementing an exact Expectation-Maximization procedure.

AB - In this paper, we propose a new model, the Kernel Kalman Filter, to perform various nonlinear time series processing. This model is based on the use of Mercer kernel functions in the framework of the Kalman Filter or Linear Dynamical Systems. Thanks to the kernel trick, all the equations involved in our model to perform filtering, smoothing and learning tasks, only require matrix algebra calculus whilst providing the ability to model complex time series. In particular, it is possible to learn dynamics from some nonlinear noisy time series implementing an exact Expectation-Maximization procedure.

U2 - 10.1109/IJCNN.2005.1556088

DO - 10.1109/IJCNN.2005.1556088

M3 - Paper

AN - SCOPUS:33750095900

SP - 1454

EP - 1459

T2 - International Joint Conference on Neural Networks, IJCNN 2005

Y2 - 31 July 2005 through 4 August 2005

ER -

Time series filtering, smoothing and learning using the kernel Kalman filter

Abstract

Conference

Access to Document

Fingerprint

Cite this