Yang Xiang

Education and Employment Background

Dr. Yang Xiang received his PhD from the University of British Columbia in 1992. In 1991, he worked as a Postdoctoral Fellow at Simon Fraser University. Following that, he held a position as a lecturer at the University of Regina, as a Visiting Professor at Aalborg University in Denmark, an Assistant Professor at the University of Regina, and as a Visiting Professor at the University of Massachusetts. Xiang joined the School of Computer Science at the University of Guelph in 2000, where he is now a Full Professor and the Associate Director of Undergraduate Studies.

Research Themes

Xiang’s research focuses on representation and inference formalism based on probabilistic and decision theoretical graphical models, such as Bayesian networks, decomposable Markov networks, influence diagrams, Markov decision processes, causal independence models, and multiply sectioned Bayesian networks. He studies issues related to representation, acquisition by machine learning or elicitation, effective inference mechanisms, and multiagent systems. He develops prototype decision support systems for various applications. Key areas of focus include:

  1. Graph-based modelling. Xiang’s research spans a wide range of issues on knowledge representation, acquisition, inference, and decision making. Emphasis is placed on understanding fundamental issues, and development of methodologies and algorithms capable of solving important classes of problems with formally provable properties. Key developments in his work include: Multiagent graphical model and algorithm for Bayesian forecasting in time series; graphical model and algorithmic framework for multiagent expedition; empirical evaluation and comparison of tightly and loosely coupled multiagent decision frameworks; partial evaluation based decision making algorithm for multiagent decision networks; empirical evaluation and comparison of multiagent constraint satisfaction algorithms; generalization of NIN-AND tree causal models to multi-valued variables; and algorithmic framework for indirect elicitation of NIN-AND tree model structures.
  2. Java-based WebWeavr toolkit. Xiang developed this toolkit, which supports graphical model-based research, teaching, and graduate student training. The toolkit is publicly available and has so far been downloaded by registered users from 30 countries.


  • Best paper award for 31st Canadian Conference on Artificial Intelligence, Canadian Artificial Intelligence Association, 2017
  • Associate Editor, Computational Intelligence, 2014-2021
  • Editorial Review Board Member, The International Journal of Data Mining, Modelling, and Management, 2008-2017
  • Editorial Review Board Member, The International Journal of Data Analysis Techniques and Strategies, 2007-2017
  • Presidential Distinguished Professor Awards, University of Guelph, 2004
  • Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery grant
  • Developer, WebWeavr, 2015