Much of the world's data is represented in terms of relations among entities. To make decisions based on such data, we want to make probabilistic predictions. These relational data are more and more being represented with rich ontologies. The ontological information is typically not reflected in the data (e.g., there is "Justin Beiber is a mammal" is not the sort of data in a dataset, because it is redundant). In this talk I will outline how relations, probabilities and ontologies can interoperate in a coherent framework. We allow probabilistic predictions about the truth of relations, properties of entities, as well as the identity, and existence of entities. Applications in geology and medicine will be presented, where we want to make predictions conditioned on all the information in the world.
David Poole is a Professor of Computer Science at the University of British Columbia. He is known for his work on combining logic and probability, assumption-based reasoning, diagnosis, relational probabilistic models, algorithms for probabilistic inference, representations and algorithms for automated decision making, probabilistic reasoning with ontologies and semantic
science. He is a co-author of two AI textbooks: "Artificial Intelligence: Foundations of Computational Agents" (Cambridge University Press, 2010, 2nd edition 2017), and "Computational Intelligence: A Logical Approach" (Oxford University Press, 1998), the book Statistical Relational Artificial Intelligence: Logic, Probability, and Computation, (Morgan & Claypool, 2016), and was co-chair of AAAI-10 and UAI-94. He is former chair of the Association for Uncertainty in Artificial Intelligence, is a Fellow of the Association for the Advancement Artificial Intelligence (AAAI) and the winner of the Canadian AI Association (CAIAC) 2013 Lifetime Achievement Award.