Fast change point detection in switching dynamics using a hidden Markov model of prediction experts

J. Kohlmorgen, S. Lemm, K. R. Mueller, S. Liehr, K. Pawelzik

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

9 Citations (Scopus)

Abstract

We present a framework for modeling switching dynamics from a time series that allows for a fast on-line detection of dynamical mode changes. The method is based on a hidden Markov model (HMM) of prediction experts. The predictors are trained by Expectation Maximization (EM) and by using an annealing schedule for the HMM state probabilities. This leads to a segmentation of the time series into different dynamical modes and a simultaneous specialization of the prediction experts on the segments. In a second step, an input-density estimator is generated for each expert. It can simply be computed from the data subset assigned to the respective expert. In conjunction with the HMM state probabilities, this allows for a very fast on-line detection of mode changes: change points are detected as soon as the incoming input data stream contains sufficient information to indicate a change in the dynamics.

Original languageEnglish
Title of host publicationIEE Conference Publication
PublisherIEE
Pages204-209
Number of pages6
Edition470
ISBN (Print)0852967217
Publication statusPublished - 1999
Externally publishedYes
EventProceedings of the 1999 the 9th International Conference on 'Artificial Neural Networks (ICANN99)' - Edinburgh, UK
Duration: 1999 Sep 71999 Sep 10

Publication series

NameIEE Conference Publication
Number470
Volume1
ISSN (Print)0537-9989

Other

OtherProceedings of the 1999 the 9th International Conference on 'Artificial Neural Networks (ICANN99)'
CityEdinburgh, UK
Period99/9/799/9/10

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Fast change point detection in switching dynamics using a hidden Markov model of prediction experts'. Together they form a unique fingerprint.

Cite this