The panel will gather together several specialists in the area of Computational Intelligence to discuss the theme of interpretability of knowledge-based systems, according to different points of view and modeling paradigms. In particular the most recent developments on interpretability-related issues such as model design, assessment and applications will be discussed. The topics covered in the panel include: formal models for interpretability, interpretable model design through CI-related approaches (not only fuzzy but also hybrid modeling techniques), interpretability assessment, and the role of interpretability in human-centric computing. The hope is to contribute in a constructive enhancement of this research direction by promoting thought-provoking discussions and cross-disciplinary integration.

Computational Intelligence or Soft Computing techniques (Fuzzy Logic, Neuro-computing, Probabilistic Reasoning, Evolutionary Computation, etc.) have already been pointed out as powerful tools for solving real-world problems when the classical methods fail. One of the main issues regarding such techniques is their cooperative nature. Each individual technique, even each individual algorithm, has its own advantages and drawbacks. Therefore, designing hybrid systems made up of different techniques working together let us achieving more powerful systems, overcoming the problems which turn up when dealing with the component techniques alone. That is why hybrid systems like for instance Neuro Fuzzy Systems (NFS) and Genetic Fuzzy Systems (GFS) are becoming more and more popular in many applications. None of them guarantee to find out the best (optimal) solution but they are able to achieve really good (sub-optimal) solutions applying approximate reasoning on the available information.

All these systems are capable of acquiring knowledge from data. However, in such cases, the problem of acquiring interpretable knowledge arises: it is important to remark that generating interpretable systems is not a straightforward task. Nevertheless, interpretability must be the central point on system modeling. In fact, some of the most hot and modern research topics like Precisiated Natural Language (PNL), Computing With Words (CWW), and/or Human Centric Computing (HCC) strongly rely on the characteristic interpretability of fuzzy models. As an example, HCC is pervading all aspects of computer science and technology. The challenge is to better exploit fuzzy logic techniques for improving the human-centric character of many intelligent systems. In humanistic systems (defined by Zadeh as those systems whose behavior is strongly influenced by human judgment, perception or emotions) readability is assumed as a prerequisite for comprehensibility.

The issue of interpretability is tackled in different ways by focusing on specific facets of system modeling, such as the formalisation of interpretability in an algebraic structure, or the definition of constraints to be imposed during model design (e.g. through genetic algorithms) and tuning (e.g. through neural networks). Finally, the evaluation of interpretability is a mandatory issue to validate and compare models. The scope of the panel is to invite experienced specialists of different areas of Computational Intelligence to discuss the interpretability of fuzzy systems with special reference to open problems and the newest research directions in all areas of Computational Intelligence.

Q1: «Why is important to keep in mind interpretability when solving real-world problems, and how much is it appreciated by users?»

Q2: «Are fuzzy systems so interpretable as people usually believe, and how to assess something so subjective as interpretability of fuzzy systems?»

Q3: «What is Human-Centric Information Processing, and how fuzzy logic can be exploited in this field?»

Q4: «Is interpretability really considered as a hot topic in the Neuro-fuzzy Computing society, and how competitive are Interpretable Neuro-fuzzy Systems?»

Q5: «New trends in Genetic and Evolutionary Fuzzy Systems: how to incorporate the new advances related to interpretability when looking for a good interpretability-accuracy trade-off in the multiobjective approach for system modeling?»

Piero Bonissone (real-world applications) General Electric Global Research, USA

Luis Magdalena (interpretability assessment) Director General at European Centre for Soft Computing, Spain

Andrzej Bargiela (Human-centric IP) Professor and member of the Automated Scheduling and Planning research group in the School of Computer Science at the University of Nottingham

Rudolf Kruse (Interpretable neuro-fuzzy systems) Professor at Otto-von-Guericke-Universität Magdeburg, Germany

Francisco Herrera (Interpretable Genetic and Evolutionary Fuzzy Systems) Professor at the Department of Computer Science and Artificial Intelligence, University of Granada, Spain

Algorithmic Trading has changed the world the way the Traders trade and Trade Support supports. There is a Brave New World happening with the "hands on" trading evolving into "hands off" algo trading. Not all trades need to be made in ultra low latency timing but the current trend has been an arms race to zero latency. Future trading will rely on a broader set of financial and economic data. Advancements will occur with Algo Trading through Computational Intelligence with intersections with Finance and Economics. The next generation of Algo Trading will encompass advanced and hybrid computational intelligence mechanisms.

Q1: «What is Computational Intelligence’s role with our next generation of Algo Trading methods?»

Q2: «What are the main issues in integrating such systems with daily workflows with finance and economics in mind?»

Q3: «Which challenges in developing effective algorithms are the hardest to overcome with Computational Intelligence methods?»

Q4: «What types of algo systems work best with which types of buy-side strategies and are they ready for the next generation of CI based Algos?»

Q5: «How should such platforms be measured and benchmarked for efficacy?»

Q6: «What are the innovations on or just beyond the horizon?»

Possible Panellists (still need to confirm)

Garnett Wilson, Afinin Labs, Canada

Daniel Paraschiv DERI/ Department of Economics National University of Ireland, Galway

Ronald R. Yager, Iona College, USA

Anthony Barabzon UCD, Dublin, Ireland

Mak Kaboudan, Professor of Statistics, School of Business, University of Redlands, USA

Moderator: Robert Golan of DBmind Technologies

There have been a growing interests in simulating brain functions at the brain scale. However, there is a large gap between connectionist modeling and symbolic modeling. Some researchers said: "Neural networks cannot do reasoning". This was negative, but it pointed out the wide knowledge gap about the brain. This situation has created a major credibility problem of all connectionist approaches to understanding the brain and to simulating the brain-like capabilities.

Connectionist approaches are bottom-up (e.g., from pixels) and symbolic approaches are top-down (e.g., from abstract concepts). Between the concrete (e.g., an edge or an edge grouping) and the abstract (e.g., goal), "much in-between" is missing. What is the "much in-between"?

This panel will bring active researchers who are interested in this subject to discuss and debate the related open questions.

This panel is co-sponsored by
IEEE CIS Autonomous Mental Development TC and its Visual Processing Task Force
INNS Autonomous Learning SIG

Q1: «Does the brain use symbolic representation in a fashion similar to our symbolic AI? Why? What lesson can we learn?»

Q2: «Connectionist models have shown progress in pattern recognition, but they have been largely used as a classifier or regressor. Some said that artificial neural networks cannot perform reasoning. Is there "a light in the tunnel" for connectionist models to perform goal-directed sensor-based reasoning? Why? What lesson can we learn?»

Q3: «Symbolic AI architectures start from abstract concepts and connectionist architectures start from concrete receptors (e.g., pixels). Hybrid architectures use both. What is the "much in-between" --- between concrete sensory inputs and abstract concepts?»

Q4: «How do you think about the brain/artificial architecture for autonomous development? As the brain is "skull-closed," how does it fully autonomously develop its internal representations for the "much in-between" through experience, from one task to the next?»

* Prof. Paolo Arena, University of Catania, Italy, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Prof. Angelo Cangelosi, University of Plymouth, UK, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Prof. Nik Kasabov, Auckland University, New Zealand, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Dr. Edgar Koerner, Honda, Germany, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Prof. Yan Meng, Stevens Institute of Technology, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Prof. Giorgio Metta, University of Genova, Italy, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Prof. Asim Roy, Arizona State University, USA, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Prof. Ron Sun, Rensselaer Polytechnic Institute, USA, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Dr. Narayan Srinivasa, HRL, USA, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Prof. Janusz Starzyk, University of Ohio, USA This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Prof. John Taylor, King’s College London, UK, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Prof. Juyang Weng, Michigan State University, USA, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

* Dr. Paul Werbos, National Science Foundation, USA, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Autonomous machine learning has become a top priority in the science and engineering of learning. In July 2007, National Science Foundation (NSF), USA, had a workshop on the "Future Challenges for the Science and Engineering of Learning". Here is the summary of the "Open Questions in Both Biological and Machine Learning" from the workshop (

 "Biological learners have the ability to learn autonomously, in an ever changing and uncertain world. This property includes the ability to generate their own supervision, select the most informative training samples, produce their own loss function, and evaluate their own performance. More importantly, it appears that biological learners can effectively produce appropriate internal representations for composable percepts -- a kind of organizational scaffold - - as part of the learning process. By contrast, virtually all current approaches to machine learning typically require a human supervisor to design the learning architecture, select the training examples, design the form of the representation of the training examples, choose the learning algorithm, set the learning parameters, decide when to stop learning, and choose the way in which the performance of the learning algorithm is evaluated. This strong dependence on human supervision is greatly retarding the development and ubiquitous deployment autonomous artificial learning systems. Although we are beginning to understand some of the learning systems used by brains, many aspects of autonomous learning have not yet been identified.”

The neural network and computational intelligence communities have a special obligation to step up to this challenge of creating autonomous learning systems that do not depend on human supervision. The International Neural Network Society (INNS) formed a Special Interest Group on Autonomous Machine Learning in 2009. This panel will be the first of a continuing series to focus on the issues raised by NSF and on the problems of creating widely deployable autonomous learning systems.

This panel will discuss our problems with creating widely deployable autonomous learning systems of various forms.

1. John Taylor - King's College, London, UK ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

2. Edgar Koerner - Honda Research Institute Europe GmbH, Germany ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

3. Klaus Obermayer - Technical University-Berlin, Germany ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

4. Jose Principe - University of Florida, Gainesville, USA ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

5. Danil Prokhorov - Toyota Research Institute, USA ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

6. Carlo Morabito - University "Mediterranea" of Reggio Calabria, Italy ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

7. Wlodzislaw Duch - Nicolaus Copernicus University, Poland ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

8. Pascal Campoy - ETSII - Universidad Politecnica Madrid, Spain ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

9. John Weng - Michigan State University, USA ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

10. DeLiang Wang - Ohio State University, USA ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

11. Fred Ham - Florida Institute of Technology, USA ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

12. Nik Kasabov - KEDRI/AUT, New Zealand ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

13. Khan M. Iftekharuddin - The University of Memphis, USA ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

14. Plamen Angelov - Lancaster University, UK ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

15. Bruno Apolloni - University of Milan, Italy ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

16. Peter Erdi - Kalamazoo College, USA ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

17. Donald Wunsch - Missouri University of Science & Technology, USA ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

18. Asim Roy - Arizona State University, USA, ( This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

This panel offers an unique opportunity to hear from both the current and immediate former editors-in-chief (EICs) of all major CIS publications about the latest news and trends in their journals. The EICs will highlight the latest developments in their journals and provide advices and suggestions on how to publish high quality papers in CIS journals, which are among the most highly rated journals in their fields in the world (according to 2008 Journal Citation Report). The panel also offers a forum for potential authors and active researchers/practitioners to exchange and explore new ideas with theEICs to further enhance the quality of CIS publications.

Q1: «What are the emphases of each CIS journal?»

Q2: «Why should I submit papers to CIS journals? What are their impact factors and circulations?»

Q3: «What is the relationship between the Magazine and Transactions?»

Q4: «What is the policy of extending a conference paper for the journal?»

Q5: «How can I get a paper accepted in a CIS journal?»

Q6: «If I have a paper cutting across two major areas covered by two journals, which one should I submit the paper to?»

Derong Liu, Editor-in-Chief of IEEE Transactions on Neural Networks

Marios Polycarpou, Former Editor-in-Chief of IEEE Transactions on Neural Networks

Nikhil Pal, Editor-in-Chief of IEEE Transactions on Fuzzy Systems

Jim Keller, Former Editor-in-Chief of IEEE Transactions on Fuzzy Systems

Garry Greenwood, Editor-in-Chief of IEEE Transactions on Evolutionary Computation

Kay Chen Tan, Editor-in-Chief of IEEE Computational Intelligence Magazine

Gary Yen, Former Editor-in-Chief of IEEE Computational Intelligence Magazine

Simon Lucas, Editor-in-Chief of IEEE Transactions on Computational Intelligence and AI in Games

Zhengyou Zhang, Editor-in-Chief of IEEE Transactions on Autonous Mental Development

Jacek Zurada, Chair of IEEE TAB Periodicals Committee

This panel discussion aims at promoting the application of computational intelligence (CI) techniques to solving complex real-world problems. Researchers from major industry will be invited to present their successful experience in pursuing CI solutions to engineering or business problems. In addition, main challenges in solving complex engineering or business problems will be discussed and potential means to deal with these difficulties will be suggested.

Q1: «What are the main promises of CI techniques for solving your problems?»

Q2: «What are the main technical and / or administrative difficulties in applying CI techniques to your problems?»

Q3: «What are your solutions to these technical and / or administrative difficulties?»

Q4: «What challenges remain to be addressed in applying CI techniques to industry / business problems?»

Q5: «What do you think can academia contribute to the solution of the remain?»