< Back to previous page

Project

Probabilistic logic programming for dynamic relational worlds

Cognitive agents need to model, reason and learn about their environments. Their environments are dynamic, i.e., they change over time, are uncertain and typically involve multiple entities and relationships. Therefore, this project focuses on dynamic relational worlds using probabilistic logic programming and statistical relational artificial intelligence principles. These are two closely related streams of research that tightly integrate probabilistic graphical models with expressive logical and/or programming language representations. The key challenges of this project include scaling up inference, learning the parameters and structure, modelling open worlds, and coping with both discrete and continuous variables. Furthermore, we will develop showcase applications in game playing and robotics.
Date:1 Oct 2018 →  30 Sep 2023
Keywords:artificial intelligence, machine learning, probabilistic reasoning, relational dynamic systems, cognitive robotics
Disciplines:Applied mathematics in specific fields, Computer architecture and networks, Distributed computing, Information sciences, Information systems, Programming languages, Scientific computing, Theoretical computer science, Visual computing, Other information and computing sciences