Analysis of nonstationary time series by mixtures of self-organizing predictors

Jens Kohlmorgen, Steven Lemm, Gunnar Raetsch, Klaus Muller

Research output: Chapter in Book/Report/Conference proceedingChapter

4 Citations (Scopus)

Abstract

We present a method for the analysis of time series from drifting or switching dynamics. In extension to existing approaches that identify switches or drifts between stationary dynamical modes, the method allows to analyze even continuously varying dynamics and can identify mixtures of more than two dynamical modes. The architecture is based on a mixture of self-organizing Nadaraya-Watson kernel estimators. The mixture model is trained by barrier optimization, a technique for constrained optimization problems. We apply the proposed method to artificially generated data and EEG recordings from the wake/sleep transition.

Original languageEnglish
Title of host publicationNeural Networks for Signal Processing - Proceedings of the IEEE Workshop
Place of PublicationPiscataway, NJ, United States
PublisherIEEE
Pages85-94
Number of pages10
Volume1
Publication statusPublished - 2000 Dec 1
Externally publishedYes
Event10th IEEE Workshop on Neural Netwoks for Signal Processing (NNSP2000) - Sydney, Australia
Duration: 2000 Dec 112000 Dec 13

Other

Other10th IEEE Workshop on Neural Netwoks for Signal Processing (NNSP2000)
CitySydney, Australia
Period00/12/1100/12/13

Fingerprint

Time series
Constrained optimization
Electroencephalography
Switches
Sleep

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Kohlmorgen, J., Lemm, S., Raetsch, G., & Muller, K. (2000). Analysis of nonstationary time series by mixtures of self-organizing predictors. In Neural Networks for Signal Processing - Proceedings of the IEEE Workshop (Vol. 1, pp. 85-94). Piscataway, NJ, United States: IEEE.

Analysis of nonstationary time series by mixtures of self-organizing predictors. / Kohlmorgen, Jens; Lemm, Steven; Raetsch, Gunnar; Muller, Klaus.

Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. Vol. 1 Piscataway, NJ, United States : IEEE, 2000. p. 85-94.

Research output: Chapter in Book/Report/Conference proceedingChapter

Kohlmorgen, J, Lemm, S, Raetsch, G & Muller, K 2000, Analysis of nonstationary time series by mixtures of self-organizing predictors. in Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. vol. 1, IEEE, Piscataway, NJ, United States, pp. 85-94, 10th IEEE Workshop on Neural Netwoks for Signal Processing (NNSP2000), Sydney, Australia, 00/12/11.
Kohlmorgen J, Lemm S, Raetsch G, Muller K. Analysis of nonstationary time series by mixtures of self-organizing predictors. In Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. Vol. 1. Piscataway, NJ, United States: IEEE. 2000. p. 85-94
Kohlmorgen, Jens ; Lemm, Steven ; Raetsch, Gunnar ; Muller, Klaus. / Analysis of nonstationary time series by mixtures of self-organizing predictors. Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. Vol. 1 Piscataway, NJ, United States : IEEE, 2000. pp. 85-94
@inbook{81c529d8a6ae453c8228d47b398fa0c6,
title = "Analysis of nonstationary time series by mixtures of self-organizing predictors",
abstract = "We present a method for the analysis of time series from drifting or switching dynamics. In extension to existing approaches that identify switches or drifts between stationary dynamical modes, the method allows to analyze even continuously varying dynamics and can identify mixtures of more than two dynamical modes. The architecture is based on a mixture of self-organizing Nadaraya-Watson kernel estimators. The mixture model is trained by barrier optimization, a technique for constrained optimization problems. We apply the proposed method to artificially generated data and EEG recordings from the wake/sleep transition.",
author = "Jens Kohlmorgen and Steven Lemm and Gunnar Raetsch and Klaus Muller",
year = "2000",
month = "12",
day = "1",
language = "English",
volume = "1",
pages = "85--94",
booktitle = "Neural Networks for Signal Processing - Proceedings of the IEEE Workshop",
publisher = "IEEE",

}

TY - CHAP

T1 - Analysis of nonstationary time series by mixtures of self-organizing predictors

AU - Kohlmorgen, Jens

AU - Lemm, Steven

AU - Raetsch, Gunnar

AU - Muller, Klaus

PY - 2000/12/1

Y1 - 2000/12/1

N2 - We present a method for the analysis of time series from drifting or switching dynamics. In extension to existing approaches that identify switches or drifts between stationary dynamical modes, the method allows to analyze even continuously varying dynamics and can identify mixtures of more than two dynamical modes. The architecture is based on a mixture of self-organizing Nadaraya-Watson kernel estimators. The mixture model is trained by barrier optimization, a technique for constrained optimization problems. We apply the proposed method to artificially generated data and EEG recordings from the wake/sleep transition.

AB - We present a method for the analysis of time series from drifting or switching dynamics. In extension to existing approaches that identify switches or drifts between stationary dynamical modes, the method allows to analyze even continuously varying dynamics and can identify mixtures of more than two dynamical modes. The architecture is based on a mixture of self-organizing Nadaraya-Watson kernel estimators. The mixture model is trained by barrier optimization, a technique for constrained optimization problems. We apply the proposed method to artificially generated data and EEG recordings from the wake/sleep transition.

UR - http://www.scopus.com/inward/record.url?scp=0034497416&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034497416&partnerID=8YFLogxK

M3 - Chapter

AN - SCOPUS:0034497416

VL - 1

SP - 85

EP - 94

BT - Neural Networks for Signal Processing - Proceedings of the IEEE Workshop

PB - IEEE

CY - Piscataway, NJ, United States

ER -