Second Workshop
The second Hybris Workshop will take place on May 27th-28th 2013 at the technical faculty of the University of Freiburg, Germany.
Venue
Click here for directions.
Program
Sunday, May 26
Travel to Freiburg
Monday, May 27
08:45 | Introduction |
09:00-09:45 | Invited talk by Robert Stevens, University of Manchester. Lessons from Teaching Non-Computer Scientists OWL and Ontologies (Abstract) |
09:45-10:15 | Talks by project partners |
B1 | Yue Ma, TU Dresden. Concept Adjustment for Formal Definitions in Description Logics |
10:15-10:45 | Coffee break |
10:45-12:45 | Talks by project partners |
B2 | Martin Gebser, University of Potsdam. Aspartame: Solving Constraint Satisfaction Problems with Answer Set Programming |
A1 | Benjamin Zarriess, TU Dresden Verification of Golog Programs over Description Logic Actions |
Martin Liebenberg, RWTH Aachen University. On Decidable Verification of Non-terminating Golog Programs | |
A2 | Hannes Strass, University of Leipzig. Abstract Dialectical Frameworks Revisited |
12:45-14:00 | Lunch |
14:00-14:45 | Invited talk by Jochen Renz, Australian National University. Hybrid Reasoning for Angry Birds (Abstract). |
14:45-15:45 | Talks by project partners |
A2 | Philipp Obermeier and Javier Davila, University of Potsdam. Progress Summary:(1) Ricochet Robots with ASP(2) A Heuristic Extension of Clasp |
C1 | Tim Niemueller, RWTH Aachen University. Towards Deliberative Active Perception - The Hybris C1 Baseline System |
15:45-16:15 | Coffee break |
16:15-16:45 | Visit of Freiburg's AI and Robotics groups. |
17:00-18:30 | Discussions between project partners. Internal Hybris meeting (PI's only). |
20:00 | Dinner for all participants |
Tuesday, May 28
Abstracts
Lessons from Teaching Non-Computer Scientists OWL and Ontologies
Invited talk by Robert Stevens, University of Manchester
Abstract: If people learning OWL are asked "from the axiom 'Person and hasPet some not Cat', How many pets can a person have?", the answer is almost invariably incorrect. OWL is a complex language, ontologies built with OWL can be complex, with complex implications. In Manchester we have been teaching OWL and ontology modelling to a broad range of people, mostly life-scientists, for over a decade. Our tutorials have been constructed to deal with what we see to be common pitfalls and mis-understandings of OWL. The tutorials are also constructed to encourage use of OWL's facilities. In biomedical ontologies we have seen a move from simple taxonomies to sophisticated use of OWL and automated reasoners. From this we can think about what it will take to make using OWL and it's facilities more accessible to a wider audience. Life-scientists have produced a large number of ontologies that have become a fantastic resource for computer scientists; by making OWL easier to use and pushing more applications of ontologies within software applications for managing and analysing data, we can create a virtuous circle of fundamentals of computer science and their application in problem areas. In this talk I will review issues in teaching and using ontologies and throw out some challenges to the community.
Hybrid Reasoning for Angry Birds
Invited talk by Jochen Renz, Australian National University
Abstract: During the IJCAI 2011 workshop on Benchmarks and Applications of Spatial Reasoning we decided that it would benefit our community to have some realistic Challenge Problems that require a substantial amount of spatial reasoning, both qualitative and quantitative. The goal was to have a way of testing the methods that have been and are being developed in a realistic setting and to identify what works and what doesn't, as well identifying gaps that cannot yet be solved using existing techniques. One of the possible problems we discussed was the popular Angry Birds game that is essentially a spatial problem.
We have now set up the Angry Birds challenge problem and provide a basic game playing framework that allows participants to focus on the underlying reasoning tasks. (seehttp://aibirds.org). We held an initial test competition in December 2012 and will have the first major competition at IJCAI 2013 this year.
While Angry Birds may appear to be a simple problem, it is actually very hard to solve for computers and consists of a number of underlying open problems from different areas of AI. I will discuss some of these problems and present in more detail one problem that cannot be solved by qualitative or quantitative reasoning alone, but requires hybrid reasoning. I will present a novel hybrid reasoning technique that is successful in solving this problem and that can be generalised to other problems.