Bayesian networks for decision-making under uncertainty
Please register for this seminar by emailing: email@example.com
Professor Ann Nicholson, Faculty of Information Technology, Monash University
Many areas of human endeavour require decision-making about complex systems, where there is: uncertainty about the causal process driving the system; limited and possibly inaccurate information about the current state of the system; and uncertainty about the effects of actions or interventions.
In this seminar, I will describe a type of computational modelling, called Bayesian decision networks, which have become a state-of-the-art technology to support decision-making under uncertainty. I will show how these models can combine data, evidence, opinion and guesstimates to help decision-makers combine probabilities and take into account costs and benefits. I will demonstrate their broad applicability through a range of examples such as biosecurity risk assessment, environmental management, fog forecasting and medical risk assessment.
Presentation slides [PDF, 9 MB]
Professor Ann Nicholson is the Deputy Dean in the Faculty of Information Technology at Monash University. After completing her BSc (Hons) and MSc in Computer Science at the University of Melbourne, in 1988 Ann was awarded a Rhodes scholarship to Oxford, where she did her doctorate in the Robotics Research Group.
After completing a post-doc at Brown University, she returned to Australia to take up a lecturing position at Monash in 1994.
Professor Nicholson specialises in the broad area of Artificial Intelligence, a sub-discipline of computer science.
Ann is a leading international researcher in the specialised area of Bayesian networks, now the dominant technology for probabilistic causal modelling in intelligent systems.
She has applied Bayesian Network (BN) technology to problem-solving in many domains including meteorology, epidemiology, medicine, education and environmental science. Examples include the use of BNs in biosecurity risk assessment, predicting the impact of conservation actions on threats and habitats of threatened species, fog forecasting and decision support for clinical cardiovascular risk assessment.