Artificial Intelligence (AI) researchers have for decades worked on building game-playing agents capable of matching wits with the strongest humans in the world, resulting in several success stories for board games like chess and checkers and computer games such as StarCraft, Pac-Man and Unreal Tournament. The success of such systems has been partly due to years of relentless knowledge-engineering effort on behalf of the program developers, manually adding application-dependent knowledge to their game-playing agents. The various algorithmic enhancements used are often highly tailored towards the game at hand.
Research into general game playing (GGP) aims at taking this approach to the next level: to build intelligent software agents that can, given the rules of any game, automatically learn a strategy for playing that game at an expert level without any human intervention. In contrast to software systems designed to play one specific game, systems capable of playing arbitrary unseen games cannot be provided with game-specific domain knowledge a priori. Instead, they must be endowed with high-level abilities to learn strategies and perform abstract reasoning. Successful realization of such programs poses many interesting research challenges for a wide variety of artificial-intelligence sub-areas including (but not limited to):
The aim of this workshop is to bring together researchers from the above sub-fields of AI to discuss how best to address the challenges of and further advance the state-of-the-art of general game-playing systems and generic artificial intelligence. Work which is appropriate for the workshop requires, but not is not limited to, work related to the General Game Playing Competition and the General Video Game Playing Competition.
The workshop is one-day long and will be held onsite at IJCAI July 10th.
The workshop papers should be submitted online (see workshop webpage). Submitted papers must adhere to the IJCAI paper formatting instructions and not exceed 8 pages (including references). The papers must present original work that has not been published elsewhere. However, submissions of papers that are under review elsewhere are allowed. All papers will be peer reviewed and non-archival working notes produced containing the papers presented at the workshop.
Important dates:
If you are interesting in attending the conference without submitting a paper please send a short statement of interest to either one of the co-chairs listed below no later than May 27th.
8:45 | Opening remarks (Michael Thielscher, Stephan Schiffel and Julian Togelius) |
9:00 | Chiara F. Sironi and Mark H. M. Winands. Optimizing Propositional Networks |
9:30 | Frédéric Koriche, Sylvain Lagrue, Éric Piette, and Sébastien Tabary. Stochastic Constraint Programming for General Game Playing with Imperfect Information |
10:00 | Bilal Kartal, Nick Sohre, and Stephen J. Guy. Generating Sokoban Puzzle Game Levels with Monte Carlo Tree Search |
10:30 | Coffee Break |
11:00 | Aline Hufschmitt and Jean-Noël Vittaut and Jean Méhat. A General Approach of Game Description Decomposition for General Game Playing |
11:30 | Stephan Schiffel. Grounding GDL Game Descriptions |
12:00 | Michael Thielscher. GDL-III: A Description Language for Epistemic General Game Playing |
12:30 | Lunch Break |
14:00 | Steven James, Benjamin Rosman, George Konidaris. An Investigation into the Effectiveness of Heavy Rollouts in UCT |
14:30 | Ivan Bravi, Ahmed Khalifa, Christoffer Holmgård, Julian Togelius. Evolving UCT Alternatives for General Video Game Playing |
15:00 | What's next for the GGP and GVGAI competitions? |
15:30 | Panel: General Game Playing and General Video Game Playing: What can we learn from each other? |
16:00 | Workshop ends |
Please help us advertise the event by circulating the CFPs.