July 17, 2006 (Monday), 3:15pm-5:15pm, Grand Ballroom AB
A Roadmap to Human-Level Intelligence

Panelists
James Anderson (Brown University)
Nick Cassimatis (Human Level Intelligence Laboratory)
Andrew Coward (Australian National University)
Kenneth De Jong (George Mason University)
Richard Duro (Universidade da Coruna)
Dario Floreano (Swiss Federal Institute of Technology)
David Fogel (Natural Selection Inc.)
Walter Freeman (University of California)
Ben Goertzel (Artificial General Intelligence Research Institute)
Steven Grossberg (Boston University)
Robert Hecht-Nielsen (University of California)
Marc de Kamps (Technical University of Munich)
Soo Young Lee (Korea Advanced Institute of Science and Technology)
John Taylor (King's College London)

Summary
Building intelligent systems with the human level of competence is the ultimate grand challenge for science and technology in general, and the computational intelligence community in particular. How are we going to achieve it? Several exciting projects aimed at reaching human-level intelligence have been formulated recently. Some of these projects start from low-level neuromorphic brain simulations, some focus on mesoscopic brain simulators, some are based on hybrid architectures and some try to develop higher-level cognitive functions at purely symbolic level. What are the merits, what are the limitations, and what can we expect at the end of each road? Potential applications span across areas of basic brain research and medicine to cognitive robotics and space research.

At the WCCI 2006 congress we plan to have a special session and a panel discussion aimed at defining a roadmap to building systems with human-level intelligence. It is a multi disciplinary subject demanding concentrated effort of experts from various fields. The emphasis will be on the scalability of the proposed models, defining the series of challenges that should be solved by these models, evolutionary and bootstrap approaches that may bring us there faster. Before the panel we shall have a special session where position papers will be presented. An edited book containing expanded versions of these papers will be published after the conference. Such organization should allow us to concentrate more on intensive discussion during the panel."

July 18, 2006 (Tuesday), 3:15pm-5:15pm, Grand Ballroom AB
Evolutionary Multi-Objective Optimization

Panelists
Carlos Fonseca (University of Algarve)
Hisao Ishubuchi (Osaka Prefecture University)
Yaochu Jin (Honda Research Institute Europe)
Joshua Knowles (University of Manchester)
Shigeru Obayashi (Tohoku University)
Patrick Siarry (University of Paris 12)
K. C. Tan (National University of Singapore)
Eckart Zitzler (ETH Zurich)

Summary
For the past 10 to 12 years, evolutionary multi-objective optimization (EMO) has received a great deal of attention among, not only the evolutionary algorithms (EAs) community, but also classical multi-criterion decision-making (MCDM) and optimization communities at large. Multi-objective optimization problems give rise to a set of Pareto-optimal solutions making a trade-off among the objectives. The main reason for the popularity and usefulness of EMO methodologies is their ability to find multiple trade-off optimal solutions in a single simulation. The range of trade-off solutions in each objective and the shape of the frontier formed by the solutions help decision-makers in making a better and more confident decision of choosing a single optimal solution. Besides, EMO methodologies have also been put to use in several other problem-solving tasks in which a consideration of additional objectives allow a more flexible and tractable search. Some EMO applications in this direction are in reducing bloating in a
genetic programming, in handling non-linear constraints, in introducing diversity in an EA, and others. EMO methodologies are also suggested to decipher useful solution principles by investigating the similarities and dissimilarities of multiple trade-off solutions.

At the WCCI 2006 congress we plan to have a special session and a panel discussion aimed at defining a roadmap to building systems with human-level intelligence. It is a multi disciplinary subject demanding concentrated effort of experts from various fields. The emphasis will be on the scalability of the proposed models, defining the series of challenges that should be solved by these models, evolutionary and bootstrap approaches that may bring us there faster. Before the panel we shall have a special session where position papers will be presented. An edited book containing expanded versions of these papers will be published after the conference. Such organization should allow us to concentrate more on intensive discussion during the panel.

Although finding a set of trade-off solutions is an important task in a multi-objective problem-solving, an equally important task is to choose a single solution with a systematic decision-making aid. In the past few years, some efforts have been spent in developing interactive EMO methodologies in which a decision-maker is factored in influencing the focus of an EMO in some preferred regions of the trade-off frontier. EMO methodologies are combined with classical MCDM techniques for this purpose. With all these developments, the research and application of EMO has become a field of its own, by attracting researchers and practitioners from various disciplines including mathematics, computer science, engineering, and commerce.

During WCCI-2006, we plan to organize a panel discussion on this important topic by addressing (i) what has been achieved so far? (ii) what are the current EMO practices? (iii) what promises do EMO provide in the short-term and long-term future? (iv) how large is the gap between academy and industry in EMO? and (v) how EMO methodologies can get benefitted from and can offer benefits to other computational intelligence techniques, such as artificial neural networks and fuzzy logic techniques. These discussions will not only allow new-comers or novices to the field to know the current EMO practices, but will also provide a platform for EMO researchers and practitioners to discuss important issues with panelists and audience. Well-renowned researchers and industrial practitioners are invited to join the panel. The panel discussion is expected to be interactive with the audience.

July 19, 2006 (Wednesday), 3:15pm-5:15pm, Grand Ballroom AB
CIS Presidents' Forum
In dedication to Walter Karplus, President of NNC, 1996

Panelists
Robert J. Marks (1990/1991) (Baylor University)
Russell Eberhart (1992/1993) (Indiana University Purdue University Indianapolis)
Patrick Simpson (1994) (Scientific Fishery Systems, Inc.)
James Bezdek (1997/1998) (University of West Florida)
Enrique Rusipini (2001) (SRI International)
Piero Bonissone (2002) (General Electric CR&D)
Evangelia Micheli-Tzanakou (2003) (Rutgers University)
Jacek Zurada (2004/2005) (University of Louisville)
Vincenzo Piuri (2006/2007) (University of Milan)

Summary
Computational Intelligence is a field that greatly evolved in the last quarter of century: from the initial steps in the direction of understanding the mechanisms of human reasoning towards the study of all aspects natural intelligence and behavior. The ultimate goal of researches in this field was mimicking nature with artificial technologies to replicate the basic mechanisms of nature in engineering systems for the benefit of the humanity. For example, understanding the brain operation is an important approach to create advanced computers with extraordinary computational power with respect to the current ones; besides understanding the learning mechanism allows for creating adaptive and evolving systems. Describing the nature by capturing the intrinsically fuzzy characteristics of our language achieves a more realistic definition of the linguistic structures, expressiveness and solutions. Evolution is useful to perform optimization and enhance solutions for better solving real problems. Cooperativeness in distributed problem solving can achieve better solutions by exploting the specialized operation and suited interactions.

Computational Intelligence technologies are not abstract ideas which have been studied only for expanding our knowledge. They are living approaches to tackle real-world problems. They have been created as answers to the needs of applications. Consequently, they are meningful and live only if they find a practical use to address problems in various application fields and to open up new fields. They have been proved useful in system modeling, control, prediction, signal processing, and image processing, data mining, for various applications encompassing, e.g., industrial automation, robotics, sensing systems, aerospace applications, automotive applications, bioengineering, biomedical applications, biometrics, bioinformatics, environmental monitoring, financial engineering, games, decision systems, and much more .

In this technological and applicative scenario the Computational Intelligence Society played a key role in promoting theoretical and applied research as well as knowledge assessment and dissemination.

•  How did CIS become the major association in the international compuational intelligence arena?
•  What CIS did to achieve this leading position?
•  Which were the main steps and achievements of CIS to support knowledge discovery in this field of interest?
•  How CIS contributed to awareness about our technologies in the various communities?
•  How did synergical cooperations with other scientific and professional associations which have common interests in some of the CIS fields helped in developing and promoting the computational intelligence?
•  How CIS was able to reach out to other scientific and professional communities for mutual benefit?
•  Which are the future directions of technological development in the area of computational intelligence?
•  Which are the aspects of nature that we feel interesting and promising for opening up new opportunities?
•  Which are the new application areas that can benefit from computational intelligence?  
•  How CIS will be able to reach out to other scientific and professional communities to expand the computational intelligence knowledge?
•  How CIS could better serve the computational intelligence community?
•  Which are the challenges that CIS like and should to take?

This panel will try to identify the achieved milestones, the learnt lessons, and the future directions of the Computational Intelligence Society for better serving our scientific and professional cummunities, in cooperation with other scientific and professional associations which are interested in some of the CIS fields .

July 20, 2006 (Thursday), 3:15pm-5:15pm, Grand Ballroom AB
Successful Real-World Applications by Computational Intelligence

Panelists
Lee A. Feldkamp (Ford Motor Company)
David Fogel (Natural Selection Inc.)
Arthur Kordon (Dow Chemical)
Leonid Perlovsky (US Air Force Research Lab)
Harold H. Szu (Office of Naval Research)
Takeshi Yamakawa (Kyushu Institute of Technology)

Summary
Since early 90´s of the 20th century, computational intelligence has been a subject of major interest for both research and applications ; among the different methodologies which are part of this field, the most well known are: evolutionary systems, fuzzy logic and neural networks. Even, if initially computational intelligence was mainly developed in research laboratories, both in the academia and in the industry, very soon it started to be applied to real-world applications. By now, it is a well established field, and even if research continues to be developed, its application is becoming more important everyday. Hence the time is adequate to disseminate the different kind of successfully applications already implemented.

At the IEEE WCCI 2006 congress we plan to have a panel session aimed at discussing these applications. The goal of this panel is to present to the audience how computational intelligence has been and continuous to be applied successfully to solve difficult real world problems almost intractable by any other technique.

Panelist will discuss a variety of applications such as: mission planning, scheduling, and drug design; detection, tracking, fusion, navigation, and sensor operations in difficult environments; applications in the chemical industry related to several key areas, such as inferential sensors, automated operating discipline, and accelerated new product development; applications in the automotive industry to relevant problems as: spark-ignition, anti-block, emission reduction; nanorobot fabrication of ultrasmall sensors; and bio-inspired chips for walking robots.

July 21, 2006 (Friday), 3:15pm-5:15pm, Grand Ballroom AB
A Practical Model for Evolutionary Computation Market Introduction

Panelists
Peter Angeline (Quantum Leap Innovations)
Carlos A. Coello Coello (CINVESTAV-IPN)
Garry Greenwood (Portland State University)
Nik Kasabov (Auckland University of Technology)
Ian Parmee (Advanced Computational Technologies)
Paul Tabor (Rizon Software LLC)
Xin Yao (The University of Birmingham)

Summary
This workshop has been running for several years and aims to assist in defining future development of the theory and application of Evolutionary Computation. During the first three years, there was a number of presentations, debates and discussions from leaders in the EC field.  These culminated in the idea records posted on the dedicated page http://www.zalzala.info/FDEC. In CEC 2005, the workshop attempted to have some means to get ideas pursuit.  It presented the model of creating a global network of academics and industrialist to redress various ideas. Rather than seeking major funding for such an initiative (which is what researchers usually do), the model simply called for creating specialized graduate programs - which academics do anyway.

This Panel Session during WCCI 2006 will present a white paper on the proposed model, from the FDEC Workgroup within CIS Evolutionary Computation Technical Committee.  It will also provide presentations on case study programs. The URL to include for the sessions is www.zalzala.info/FDEC