Lifted Dynamic Junction Tree Algorithm

For relational temporal models, the Lifted Junction Tree algorithm efficiently answers filtering and predication requests by building a parameterized grouping representation from a knowledge base and reusing it for many queries and time steps.

Marcel Gehrke gives a talk on the paper "Lifted Dynamic Junction Tree Algorithm" Marcel Gehrke, Tanya Braun, and Ralf Möller at the 23rd International Conference on Conceptual Structures (ICCS) from 20.-22. June 2018 in Edinburgh.