Efficient inference in dynamic belief networks with variable temporal resolution

Timothy A Wilkin, Ann E Nicholson

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    2 Citations (Scopus)


    Dynamic Belief Networks (DBNs) have been used for the monitoring and control of stochastic dynamical processes where it is crucial to provide a response in real-time. DBN transition functions are typically specified as conditional probability distributions over a constant time interval. When these functions are used to model dynamic systems with observations that occur at irregular intervals, both exact and approximate DBN inference algorithms are inefficient. This is because the computation of the posterior distribution at an arbitrary time in the future involves repeated application of the fixed time transition model. We draw on research from mathematics and theoretical physics that shows the dynamics inherent to a Markov model can be described as a diffusion process. These systems can be modelled using the Fokker-Planck equation, the solutions of which are the transition functions of the system for arbitrary length time intervals. We show that using these transition functions in a DBN inference algorithm gives significant computational savings compared to the traditional constant time-step model.
    Original languageEnglish
    Title of host publicationPRICAI 2000 Topics in Artificial Intelligence
    Subtitle of host publication6th Pacific Rim International Conference on Artificial Intelligence Melbourne, Australia, August 28 - September 1,2000 Proceedings
    EditorsRiichiro Mizoguchi, John Slaney
    Place of PublicationBerlin Germany
    Number of pages11
    ISBN (Print)3540679251
    Publication statusPublished - 2000
    EventPacific Rim International Conference on Artificial Intelligence 2000 - Melbourne, Australia
    Duration: 28 Aug 20001 Sept 2000
    Conference number: 6th
    https://link-springer-com.ezproxy.lib.monash.edu.au/book/10.1007/3-540-44533-1 (Proceedings)

    Publication series

    NameLecture Notes in Computer Science
    ISSN (Print)0302-9743


    ConferencePacific Rim International Conference on Artificial Intelligence 2000
    Abbreviated titlePRICAI 2000
    Internet address

    Cite this