Recent years have seen a surge of interest in combining logic with probabilities. This year, the international workshop on “Statistical Relational Learning” (SRL-09) was again highly successful. One of the interesting take-away messages is that many researchers would like to go one step ahead: applying lifted inference and SRL methods to general AI tasks. There is, for example, evidence that exploiting shared factors within message-passing algorithms can greatly speed up probabilistic relational inference but also SAT-like problems. Motivated by this, we are organizing the
One of the goals of the workshop is to form a common core of problems and ideas across all the different AI subfields such as knowledge representation, vision, planning, SAT etc. So far, researchers combining logic and probability in each of these subfields have been working mostly independently. We believe the current situation actually provides us with an opportunity for attempts at synthesis, forming a common core of problems and ideas, and cross-pollinating across subareas. We would like to explore the minimal perturbations required for each of the AI subfields to start using statistical relational (SR) techniques.
We seek to invite researchers in all subfields of AI to attend the workshop and to explore together how to reach the goals imagined by the early AI pioneers.
* We thank Pedro Domingos for his valuable suggestions and discussions.
Update (23 Apr 2010): The list of accepted papers has been posted.
Update (29 Mar 2010): Due to the large number of requests, the deadline for submission has been extended to March 31, 2010.
Update (22 Mar 2010): The submission site is open.
Update (24 Nov 2009): The call for papers is out.
NIPS 2008 Workshop on Probabilistic Programming
Dagstuhl Seminar on Probabilistic, Logical and Relational Learning 2007
Dagstuhl Seminar on Probabilistic, Logical and Relational Learning - Towards a Synthesis 2005
The very first workshop on SRL - 2000
Neural-Symbolic Learning and Reasoning 2010