In many real-world applications, especially in Artificial Intelligence, agents need to make decisions or to optimize functions under uncertain information. Successes in such applications stimulated the need to handle problems of ever increasing size and, by their decomposition ability, Graphical Models naturally imposed themselves as one of the most popular modeling and reasoning tool. The objective of this special track on “Graphical Models: from Theory to Applications” is to bring together researchers working on these models or exploiting them in their applications in order to share their experience and discuss the latest developments in this field. Graphical Models of interest for this special track include but are not limited to Bayesian networks, Markov random fields, credal networks, influence diagrams, factored Markov decision processes, sum-product networks, factor graphs. Prospective authors are invited to submit original papers to this Special Track.