Call For Papers
Accepted Papers
Program Committee
Local Information
Important Dates



IEEE CIG 2007 Keynote Talks

Michael Greenspan: Deep Green: Playing Pool Against a Robot

Michael Buro: Game AI Trends and Challenges

Marco Ernandes: The WebCrow Man-Machine Crossword Challenge

Michael Greenspan: Deep Green: Playing Pool Against a Robot

Since the seminal work on computational chess of Claude Shannon and Alan Turing, computer gaming has had a long and respected tradition, and the success in 1997 of IBM’s Deep Blue against world chess champion Gary Kasparov is regarded as a turning point in the evolution of computational intelligence. To date, however, no robotic system has successfully competed against a proficient human opponent. This talk will describe the development of Deep Green, a Robotic Intelligent System being developed to compete against humans at the game of pool, which currently plays 8 Ball at a better-than-amateur level. The Deep Green system comprises: a ceiling mounted gantry robot; a cue end-effector; a machine vision system; control and physical modeling algorithms; a search-tree based strategy engine; and a full-sized pool table. The background research and major design decisions will be reviewed, and recent results in automatic play will be presented. The talk will conclude with a description of the future research problems that will be addressed to advance the system to perform at a competitive level.


Michael Greenspan is an Associate Professor with the Department of Electrical and Computer Engineering and the School of Computing (cross-appointed) at Queen’s University, Kingston, Canada. He was awarded a B.Sc. in Physics and Applied Mathematics from the University of Toronto, a B.A.Sc. and M.A.Sc. in Electrical Engineering from the University of Ottawa, and a Ph.D. from Carleton University, Department of Systems and Computer Engineering, in 1999. From 1991 to 2001, he was employed by the Institute for Information Technology of the National Research Council of Canada, acting as a researcher and ultimately as the group leader of the Computational Video Group. Dr. Greenspan holds membership with the Professional Engineers of Ontario, the IEEE Computer Society, the Research Management Committee of Precarn Associates, and the Canadian Image Processing and Pattern Recognition Society (CIPPRS). Dr. Greenspan was the recipient of the CIPPRS 2003 Young Investigators Award, the Premier’s Research Excellence Award, and was named the Favourite Second Year Professor by the Queen’s ECE Club in 2003 and 2004. His research interests include Computer Vision, Object and Pattern Recognition, Robotics, and Computer Gaming.

Game AI Trends and Challenges

Michael Buro

Some of the major accomplishments of AI research has been the creation of programs challenging human World Champions in classic games, such as chess, checkers, Othello, Scrabble, and Backgammon. In recent years we have started to see progress in more challenging game AI domains, like modern video games, as well in terms of machine learning and planning. In this presentation I will give an overview of techniques that have the potential to revolutionize commercial game AI. I will also argue that the video games industry needs to reconsider its limiting view of AI - as merely having to create the illusion of intelligence - by addressing the real problem, namely creating genuinely intelligent game characters.

The WebCrow Man-Machine Crossword Challenge

Authors: Giovanni Angelini, Marco Ernandes, and Marco Gori

Speaker: Marco Ernandes

Solving crossword puzzles requires a wide knowledge in different domains and the ability to crack enigmatic clues, that are often regarded as inherent human capabilities. Unlike most games faced by computers, crossword solving is not difficult because of the need to perform sophisticated strategic plans in a huge yet finite universe of configurations. Crossword solving, which has been informally classified as an AI-complete problem, is difficult to attack because it requires to capture the linguistic specifications and, consequently, to discover appropriate answers.

In this talk we will discuss the problem of automatic crossword solving with special emphasis on WebCrow, a project carried out at the University of Siena, which is conceived to attack crosswords in different languages. We present the main problems to be faced and provide some relevant architectural issues behind cracking crosswords. In particular, we will focus attention on the crucial role of the Web clue-answering module, which uses a number of computational intelligence techniques, including machine learning-based clue classification to improve the ranking of the candidate answers, that are subsequently allocated by a constraint satisfaction engine. Finally, we will report the results that we have already collected on a number of man-machine competitions for American, Italian, and bi-lingual crosswords. Moreover, we will also provide an update with the results of an on-line American crossword competition that will take place immediately before the conference.