Details of Grant 

EPSRC Reference:	GR/K72254/01
Title:	DYNAMIC PROBABILISTIC EXPERT SYSTEMS
Principal Investigator:	Smith, Professor JQ
Other Investigators:
Researcher Co-Investigators:
Project Partners:
Department:	Statistics
Organisation:	University of Warwick
Scheme:	Standard Research (Pre-FEC)
Starts:	01 September 1996	Ends:	31 August 1999	Value (£):	124,025
EPSRC Research Topic Classifications:	Statistics & Appl. Probability
EPSRC Industrial Sector Classifications:
Related Grants:
Panel History:
Summary on Grant Application Form
Despite early scepticism it has now been established that probabilistic expert systems can be built and these are not only faithful and exact methodologies but are also quick. They are based on a graphical framework called an influence diagram. However, the current situations in which these methodologies can be applied are somewhat limited. The current algorithms generally work when it is possible to specify a fixed number of variables whose joint distribution can be represented in a particular way - it needs to be decomposable and its clique sizes need to be small. Furthermore unless numerical techniques are employed, the user needs to make strong distributional assumptions about these variables. In this research I plan to develop methods of efficient probability propagation which address these constraints on the field of application of probabilistic systems. These will code uncertainties and provide the framework and algorithms for propagating probabilities in a setting where: (i) new variables are being introduced continually into the system; (ii) the joint distributions are not necessarily decomposable; (iii) there are much lighter restrictions on the distributional forms of the random variables concerned.The methods I intend to use first define a dynamic influence diagram and then use dynamic approximation techniques (Solution Track 1), alternative conditional representations of probability distributions (Solution Track 2) and efficacious transformations (Solution Track 3) to improve on and widen the field of application of the current propagation methodologies.
Key Findings
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Potential use in non-academic contexts
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Impacts
Description This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk Summary Date Materialised
Sectors submitted by the Researcher
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Project URL:
Further Information:
Organisation Website:	http://www.warwick.ac.uk