CALL FOR CONTRIBUTIONS
International Workshop on Statistical Relational Learning (SRL-2009)
2-4 July 2009
AIMS AND SCOPE
The SRL workshop series started in 2000 and is the premier international forum on statistical relational learning. The SRL 2009 workshop will take place in Leuven (Belgium) from 2-4 July (which is just after SIGKDD in Paris, with Leuven being only 2 hours away by train from Paris). SRL-2009 will be co-located with
- ILP 2009, the 19th International Conference on Inductive Logic Programming
- MLG 2009, the 7th International Workshop on Mining and Learning with Graphs
and will feature joint invited speakers and tutorials, paper presentations as well as poster sessions. It is the first time that these three meetings on neighboring fields are organized together with the aim of stimulating discussion and interaction across the involved disciplines and, hence, providing a complete view on analyzing structured data.
CALL FOR CONTRIBUTIONS
Statistical relational learning (SRL) deals with machine learning in relational domains where observations can be missing, partially observed, and/or noisy. It thus addresses the challenge of applying statistical inference and learning approaches to problems which involve rich collections of objects linked together in complex relational networks. It is the connections among objects that are of crucial importance as information about entities in isolation is often limited. Application areas include biology, robotics, ubiquitous computing, social network analysis, among others.
In the tradition of the highly successful SRL workshops and Dagstuhl seminars in the past, SRL-2009 again will be the premier venue for bringing together different sub-disciplines that focus on the probabilistic analysis of relational data. So far, the SRL research has already provided many novel insights and useful approaches. Still, there are many open problems and challenges lying ahead. How can we apply SRL in the wild? How do we lift inference? How do we deal with changing environments? What are the next killer applications? What lessons can SRL, MLG, and ILP learn from each other?
Therefore, we would like to encourage submissions to the following (not exclusive) list of topics that are centered around theoretical, algorithmic, and implementional aspects of SRL:
- sample complexity results
- probabilistic programming
- mapping between different SRL approaches
- unifying approaches
- parameter estimation and model selection approaches
- lifted inference
- statistical predicate invention
- nonparametric (Bayesian) approaches
- relational decision theoretic planning and reinforcement learning
- learning complex actions
- semi-supervised relational learning
- preference learning
- clustering relational data
- systems and toolboxes
Last but not least, we would like to especially encourage people to submit CHALLENGE papers, describing one or more technical challenges for the SRL community, as well as APPLICATION papers, describing show-case applications of SRL.
Submissions take the form of extended abstracts (at most 3 pages following ICML style). The abstract should include a title, the authors' names as well as electronic and physical addresses. In addition to original contributions, submissions based on papers that have recently been published in good venues will be considered, given that the original papers are adequately cited and their status is clearly stated. Since the workshop wants to present the most recent and best research results in the field, multiple submissions are allowed.
All abstracts will be reviewed by at least three referees. Reviewing will not be double-blind. On the basis of the reviews, it will be decided whether the abstract is accepted for oral presentation, for presentation as a poster or rejected.
A joint special issue of the Machine Learning Journal on ILP - SRL - MLG will be published after the conference. Authors of selected papers will be invited after the conference to submit a longer version (up to 20 pages) of their paper to the special issue by September 1. These papers will be on a fast track to publication in the issue, which implies that decisions on whether or not to accept the papers (possibly with revisions) will be made quickly.
- April 3 Deadline for paper / abstract submissions
(updates to submitted papers will remain possible until April 7)
- April 30 Notification
- May 29 Deadline for camera ready copies
- July 2-4 Workshop(s) / Conference(s)