| IEEE HOME | IEEE WEB ACCOUNTS | IEEE XPLORE | SMC SOCIETY | CONTACT IEEE |
IEEE Office 3 Park Avenue, 17th Floor New York, New York 10016-5997 U.S.A. Tel: +1 212 419 7900 Fax: +1 212 752 4929
President's Message Editorial News Focus on Technical Committees From Society Leaders Submission Guidelines Back Issues Society Information

Focus on Technical Committees


The Systems, Man, and Cybernetics Society is home to three technical groups: Cybernetics, Human Machines Systems, and System Science and Engineering. The Cybernetics group includes 16 committees, the Human Machine Systems, 8 TC's, and the largest is System Science and Engineering with 18 committees. In each issue of this Newsletter, we cover the activities of some technical committees to familiarize the reader with the committee's work and stimulate interest in their activities.

Technical Committee on Evolving Intelligent Systems

Communicated by Palmen Angelov


Cybernetics is a mature research area and to some degree this is true for the intelligent systems research area which can be seen as much closer to the real life and challenges, more engineering discipline as compared to the AI. In both areas, however, the problems of adaptation and self-learning, dynamic changes of the (often complex and unpredictable or even hostile) environment and systems of interest has either be considered in very limited, idealised scenarios (such as linear or simplified quadratic models) or ignored all together assuming off-line scenarios, ideal distributions, availability of all data in advance, etc.

The research area of so called 'evolving systems' and EIS in particular emerged as a more systematic discipline and niche area with small but growing and energetic community of very active researchers addressing this need to bridge the gap between the real life problems in various areas such as process industries (chemical, petrochemical), defence, security, autonomous systems, ambient assisted living, ubiquitous computing, mobile robotics, emotive intelligence, advanced car and aviation industry, cyber security, etc. on one side and the limitations of the state of the art methodologies and concepts.

One of the important research challenges today is to design intelligent systems that have a higher level of flexibility and autonomy that can develop their understanding of the environment and ultimately their intelligence. It is to be noted that the environments in which such systems are required to successfully operate are very often ill-defined - they are non-stationary, (often unpredictably) changing, partially or completely unknown. To address the problems of modelling, control, prediction, classification, clustering, anomaly detection and data processing in such environments a system must be able to fully adapt, not simply to adjust parameters of a pre-trained and fixed structure. That is, the system must be able to evolve, to self-develop, and to self-organize.

While conventional adaptive techniques are suitable to represent objects with slowly changing parameters, they can hardly handle complex systems with multiple operating modes or abruptly changing characteristics since it takes a long time after every drastic change in the system to relearn model parameters.  The evolving systems paradigm is based on the concept of evolving (expanding or shrinking) model structure that is capable of adjusting to the changes in the objects that cannot solely be represented by parameter adaptation.

The emerging area of evolving intelligent systems targets non-stationary processes by developing novel on-line learning methods and computationally efficient algorithms for real-time applications. Evolving intelligent systems are evolution-inspired. They focus on the evolution of an individual system rather than of a population as the conventional genetic (evolutionary) algorithms do.

They use inheritance and gradual change with the aim of life-long learning and adaptation, self-organization (including system structure evolution) in order to adapt to the (unknown and unpredictable) environment. This concept is applicable to a wide range of systems - probabilistic (evolvable Bayesian classifiers, decision trees), fuzzy, neural networks, languages. Cognitive and psychological aspects of the evolution of individual systems will also be of interest.

The TC on EIS (web site http://www.ieeesmc.org/technicalcommittess/tc_evis.html ) currently includes 21 members from all continents, genders and 14 countries who are active researchers in this emerging area. It was established in 2011 and coordinated a long list of activities including a regular annual conference IEEE Conference on Evolving and Adaptive Intelligent Systems - fully sponsored by the SMC Society (the last one being in 2012 in Madrid), Summer Schools, special sessions, books and tutorials, etc.

The goals for the next five years include increasing the number of members, consolidating the conference and other events (summer school, special sessions, tutorials, books as well as new initiatives, such as workshops, competitions, special issues in the Magazines and prestigious journals).

Some of the symbolic/important papers since the committee started include:

1. Federico Montesino Pouzols and Amaury Lendasse, Evolving fuzzy optimally pruned extreme learning machine for regression problems
2. Abdelhamid Bouchachia, An evolving classification cascade with self-learning
3. Ana Belen Cara, Hector Pomares, Ignacio Rojas, Zsofia Lendek and Robert Babuska, Online self-evolving fuzzy controller with global learning capabilities



Human Perception in Vision, Graphics and Multimedia Committee

The committee was established in January 2010 under the leadership of co-chairs: (from left to right above) Anup Basu, and Irene Cheng, University of  Alberta, and Holly Rushmeier, Yale University. The Industrial Co-chair is Haohong Wang, TCL Research.

Topics within the scope of the TC are: Perceptual quality in image, video and 3D compression; QoE (Quality of Experience) based Multimedia and Graphics; Perceptual factors & QoE in DASH (Dynamic Adaptive Streaming over HTTP); 3D mesh refinement and evaluation of simplification algorithms based on perceptual thresholds and skeletonization; Human factors in 3DTV and stereo visualization; Visual quality prediction and perceptually driven texture reduction; Active Vision; Panoramic view perception, capture and rendering; Perceptually optimized transmission of integrated texture and mesh taking packet loss into consideration; Foveation for efficient image, video and 3D transmission;  and, Perceptual factors in efficient web-based multimedia education.

Our goals for the next five years include: Organizing the 2013 IEEE International Conference on Multimedia and Expo (ICME 2013) in San Jose; Organizing the Industrial Track of IEEE SMC 2014 conference in San Diego; Organizing Special Sessions at IEEE SMC Conferences; Organizing the STAR Report presentation at the Eurographics conference in 2012; Organizing Special Issues on HMS in IEEE SMC Transactions; and Organizing Panels to Enhance University-Industry Interactions .

Three of our most important papers are:

1. "Perceptually Coded Transmission for Arbitrary 3D Objects over Burst Packet Loss Channels and a Generic JND Formulation," (I. Cheng, L. Ying and A. Basu), IEEE Journal on Selected Areas of Communication (JSAC), 10 pages, accepted March 2012.

2. "Optimal Pixel Aspect Ratio for Enhanced 3D TV Visualization," (H. Azari, I. Cheng and A. Basu), Computer Vision, Graphics and Image Processing, 38-53, Jan 2012.

3. "Perceptually Guided Fast Compression of 3D Motion Capture Data," (A. Firouzmanesh, I. Cheng and A. Basu), IEEE Transactions on Multimedia, 829-834, Aug. 2011.