CIG'08
IEEE Symposium on Computational Intelligence and Games

Perth, Australia
15 - 18 December 2008

The 2008 CIG symposium will host a number of special sessions that examine exciting new areas of research in computational intelligence in games. Special sessions provide participants with opportunities to discuss and explore these emerging areas with fellow researchers and practitioners in a focussed session that examines the specialised area in significant detail. These special sessions provide a means of shaping the overall direction of the symposium, and are a vital mechanism for theming submitted works into cognizant areas of interest.

Special sessions will be scheduled along with technical sessions. Attendance at all special sessions is included in the symposium registration fee.

Special Session 1: Computational Intelligence in Real Time Strategy Games
Chairs: Mike Preuss, Bobby Bryant and Chris Miles
Contact: mike.preuss AT uni-dortmund.de

Real Time Strategy Games (RTSG) are one of the major branches of the current industrial game production. They often lure the player into another (e.g. historically motivated) world and usually assign him the task of leading a faction into a dominating state by exerting total (omnipotent) control over all buildings, units, and policies of his realm. However, it seems that the artificial intelligence controlling the adversaries did not develop on the same scale as the multimedia features (graphics, sound) of the games, often leading to situations that can be exploited by experienced gamers. We presume that CI is a possible way out here, providing more realistic and challenging game behavior. Additionally, CI techniques may also be helpful in supporting the human player's micro-management, thereby paving the way for handling more complex game features. Nevertheless, not much work has been done in this area until now, so it is not at all clear if CI techniques or other (e.g. machine learning) methods are better suited here.
The major aim of the special session is to identify and collect problems/shortcomings in currently existing RTSGs and their AIs, and to evaluate the potential of different CI methods for improving game behavior. Possibly rewarding fields for applying CI techniques could also be game design/balancing (e.g. unit strengths), support for micro-management, and control of player-supporting non-player characters (NPCs). As well as feasibility studies, it shall also be evaluated how robust the methods are, how they compare with alternative approaches, and how difficult it is to apply them.
We will also dedicate some time to discussion of the potential of CI methods in RTSGs, as well as of current and future trends in RTSGs.

Mike Preuss is Research Associate at the Computer Science Department, Dortmund University of Technology, Germany (since 2000), where he also received his diploma degree in 1998. His current research interests focus on the field of evolutionary algorithms and include niching, multiobjective optimization, games and other real-world applications. He focuses on an experimental approach to algorithm design and analysis and on developing the experimental methodology in CI. Games, and especially RTSG, are of specific interest to him as creating AI players for them poses a major challenge (multiple objectives, incomplete knowledge, inherent randomness...) for modern CI methods.
Bobby Bryant is an Assistant Professor of Computer Science and Engineering at the University of Nevada, Reno. He has been using a discrete state strategy game to study intelligent agents since 2003, addressing such problems as in situ divisions of labor, policy induction from examples, and visibly intelligent (as opposed to strictly optimal) behavior. He is now the chair for a task force on RTSGs for the IEEE Computational Intelligence Society's Games Technical Committee, and directs the Neuroevolution and Behavior Laboratory at UNR. His homepage can be found at: http://www.cse.unr.edu/~bdbryant/.

Special Session 2: Player Satisfaction
Chairs: Georgios Yannakakis and Bobby Bryant
Contact: yannakakis AT itu.dk

The special session on Player Satisfaction will be the third event in a series started with two successful workshops in the field: the first workshop on Adaptive Approaches for Optimizing Player Satisfaction in Computer and Physical Games in conjunction with the Simulation of Adaptive Behavior (SAB) conference in 2006 and the workshop on Optimizing Player Satisfaction held on July 6-8, 2007, in conjunction with the Artificial Intelligence and Interactive Digital Entertainment (AIIDE-07) conference sponsored by AAAI.
As in our first two workshops, our objective for holding this special session at CIG'08 is to initiate/boost the game AI community's and game developers' interest on the study, development and evaluation of methodologies for modeling and augmenting player satisfaction. An additional goal of the event is to encourage a dialogue among researchers in AI, human-computer interaction, affective computing and psychology disciplines who investigate dissimilar CI methodologies for improving game-play experiences. Furthermore, we expect this event to yield a better understanding of state-of-the-art approaches for improving player satisfaction in games. Research areas relevant to Player Satisfaction include, but are not limited to, the following:

  • Player Satisfaction Modeling:
    • Empirical Models
    • Psychological Models
    • Cognitive Models
    • Affective Models
  • Optimizing Player Satisfaction:
    • CI approaches
    • Adaptive learning
The organisers plan to invite a recognized game developer/game designer who will place Player Satisfaction under the industrial viewpoint.

Georgios Yannakakis is Assistant Professor at the Center for Computer Games Research, IT-University of Copenhagen. His primary research interests lie in cognitive (entertainment) modeling, affective computing, neuro-evolution, dynamic game balancing, and real-time learning in video games. Dr. Yannakakis is the chair of the IEEE CIS Task Force on Player Satisfaction Modeling. He has been the chair of two workshops (SAB'06, AIIDE'07) on areas strongly related to player satisfaction in games and program committee member of several game related conferences and workshops. His homepage can be found at: http://www.itu.dk/~yannakakis/.
Bobby Bryant is an Assistant Professor of Computer Science and Engineering at the University of Nevada, Reno. He has been using a discrete state strategy game to study intelligent agents since 2003, addressing such problems as in situ divisions of labor, policy induction from examples, and visibly intelligent (as opposed to strictly optimal) behavior. He is now the chair for a task force on RTSGs for the IEEE Computational Intelligence Society's Games Technical Committee, and directs the Neuroevolution and Behavior Laboratory at UNR. His homepage can be found at: http://www.cse.unr.edu/~bdbryant/.

Special Session 3: Coevolution in Games
Chairs: Julian Togelius, Alan Blair and Philip Hingston
Contact: julian.togelius AT gmail.com

In coevolution, the fitness of a solution is determined not (only) by a fixed fitness function, but also by the other solution(s) being evaluated. Thus, coevolution has the potential to overcome several problems with static fitness functions, paving the way for more open-ended evolution. However, several phenomena common to coevolutionary algorithms are at present poorly understood, including cycling and loss of gradient. Further understanding of such phenomena would facilitate more widespread use of coevolutionary algorithms.
This special session seeks to bring together research that uses coevolutionary algorithms to learn to play games, uses games to investigate coevolution, or uses coevolution as a basis for game design. Due to their adversarial nature, often involving interaction of multiple agents, games are uniquely suited to be combined with coevolution. We invite both theoretical and applied work in the intersection of coevolution and games, including but not limited to the following topics:
  • Competitive coevolution
  • Cooperative coevolution
  • Multiple populations in coevolution
  • Coevolution with diverse representations
  • Theory of coevolution
  • Preventing cycling and loss of gradient
  • Coevolution-based game design
  • Self-play and coevolutionary-like reinforcement learning
  • Relative versus absolute fitness metrics

Julian Togelius is a researcher at the Dalle Molle Institute for Artificial Intelligence (IDSIA) in Lugano, Switzerland. His research interests include evolving game-playing agents, modelling player behaviour, and evolving interesting game content, mainly using evolutionary and coevolutionary techniques. He also co-organizes the well-attended Simulated Car Racing Competitions for the IEEE CIG and CEC conferences. Julian's homepage is available from: http://julian.togelius.com/.
Alan Blair is Chair of the IEEE CIS Task Force on Co-evolution and Games. His research interests include robot navigation, image and language processing as well as co-evolutionary learning for Backgammon, Tron, IPD, simulated hockey and language games. His homepage is at: http://www.cse.unsw.edu.au/~blair.
Philip Hingston is an associate professor of computer science at Edith Cowan University in Perth. His research interests are in the theory and application of artificial intelligence and computational intelligence. He has a particular interest in evolutionary computation as a tool for design, and in computer games. He is chair of the IEEE CIS Task Force on co-evolution. More information can be found on his home page at: http://www.scis.ecu.edu.au/Staff/staffinfo.aspx?staffid=phingsto.

Special Session 4: Player/Opponent Modeling
Chairs: Bobby Bryant and Philip Hingston
Contact: bdbryant AT cse.unr.edu

Behavior Cloning, Policy Induction, Behavior Capture, Strategy Capture, Style Machines - regardless of the name, the goal is the same: to create a model of an agent on the basis of examples or observations. In games and simulations such techniques are useful for opponent modeling - creating a model of an opponent so you can adapt to beat it - and for player modeling - creating a model of a player so you can can adapt to beat or aid the player, or tune the game to a player-specific focus of interest or difficulty level.
This special session will bring together papers on techniques and applications for deriving agent models from observation or interaction in game and simulation environments, plus papers on methods for evaluating the accuracy of such derived models. The modeled agents can be players or software agents that play games, or active situated agents that live in a game or game-like environment.

Bobby Bryant is an Assistant Professor of Computer Science and Engineering at the University of Nevada, Reno. He has been using a discrete state strategy game to study intelligent agents since 2003, addressing such problems as in situ divisions of labor, policy induction from examples, and visibly intelligent (as opposed to strictly optimal) behavior. He is now the chair for a task force on RTSGs for the IEEE Computational Intelligence Society's Games Technical Committee, and directs the Neuroevolution and Behavior Laboratory at UNR. His homepage can be found at: http://www.cse.unr.edu/~bdbryant/.
Philip Hingston is an associate professor of computer science at Edith Cowan University in Perth. His research interests are in the theory and application of artificial intelligence and computational intelligence. He has a particular interest in evolutionary computation as a tool for design, and in computer games. He is chair of the IEEE CIS Task Force on co-evolution. More information can be found on his home page at: http://www.scis.ecu.edu.au/Staff/staffinfo.aspx?staffid=phingsto.