Keynote Lectures List:
- Daniel S. Yeung, University of Technology, China
Title: Sensitivity Based Generalization Error for Supervised Learning Problem with Applications in
Model Selection and Feature Selection

- Maria P. Fanti, Polytechnic of Bari, Italy
Title: Impact of the ICT on the Management and Performance of Intelligent Transportation Systems

- Janan Zaytoon, University of Reims Champagne Ardennes, France
Title: Recent Advances in Verification and Analysis of Hybrid Systems

- Alessandro Giua, Universita' di Cagliari, Italy
Title: Discrete Event Diagnosis using Petri Nets

- Peter S. Sapaty, Institute of Mathematical Machines and Systems, National Academy of Sciences Ukraine
Title: Meeting the World Challenges: From Philosophy to Information Technology to Applications
Keynote Lecture 1
Sensitivity Based Generalization Error for Supervised Learning Problem
with Applications in Model Selection and Feature Selection
  Daniel S. Yeung,
University of Technology
Brief Bio
Daniel S. Yeung (Ph.D., M.Sc., M.B.A., M.S., M.A., B.A.) is the President of the IEEE Systems, Man and Cybernetics (SMC) Society, a Fellow of the IEEE and an IEEE Distinguished Lecturer. He received the Ph.D. degree in applied mathematics from Case Western Reserve University. In the past, he has worked as an Assistant Professor of Mathematics and Computer Science at Rochester Institute of Technology, as a Research Scientist in the General Electric Corporate Research Center, and as a System Integration Engineer at TRW, all in the United States. He was the chairman of the department of Computing, The Hong Kong Polytechnic University, Hong Kong, and a Chair Professor from 1999 to 2006. Currently he is a Chair Professor in the School of Computer Science and Engineering, South China University of Technology, Guangzhou, China.
His current research interests include neural-network sensitivity analysis, data mining, Chinese computing, and fuzzy systems. He was the Chairman of IEEE Hong Kong Computer Chapter (91and 92), an associate editor for both IEEE Transactions on Neural Networks and IEEE Transactions on SMC (Part B), and for the International Journal on Wavelet and Multiresolution Processing. He has served as a member of the Board of Governors, Vice President for Technical Activities, and Vice President for Long Range Planning and Finance for the IEEE SMC Society. He co-founded and served as a General Co-Chair since 2002 for the International Conference on Machine Learning and Cybernetics held annually in China. He also serves as a General Co-Chair (Technical Program) of the 2006 International Conference on Pattern Recognition. He is also the founding Chairman of the IEEE SMC Hong Kong Chapter.
His past teaching and academic administrative positions include a Chair Professor and Head at Department of Computing, The Hong Kong Polytechnic University, the Head of the Management Information Unit at the Hong Kong Polytechnic University, Associate Head/Principal Lecturer at the Department of Computer Science, City Polytechnic of Hong Kong, a tenured Assistant Professor at the School of Computer Science and Technology and an Assistant Professor at the Department of Mathematics, both at Rochester Institute of Technology, Rochester, New York.
He also held industrial and business positions as a Technical Specialist/Application Software Group Leader at the Computer Consoles, Inc., Rochester, New York, an Information Resource Sub-manager/Staff Engineer at the Military and Avionics Division, TRW Inc., San Diego, California, and an Information Scientist of the Information System Operation Lab, General Electric Corporate Research and Development Centre, Schenectady, New York.

Generalization error model provides a theoretical support for a classifier's performance in terms of prediction accuracy. However, existing models give very loose error bounds. This explains why classification systems generally rely on experimental validation for their claims on prediction accuracy. In this talk we will revisit this problem and explore the idea of developing a new generalization error model based on the assumption that only prediction accuracy on unseen points in a neighbourhood of a training point will be considered, since it will be unreasonable to require a classifier to accurately predict unseen points "far away" from training samples. The new error model makes use of the concept of sensitivity measure for an ensemble of multiplayer feedforward neural networks (Multilayer Perceptrons or Radial Basis Function Neural Networks). Two important applications will be demonstrated, model selection and feature reduction for RBFNN classifiers. A number of experimental results using datasets such as the UCI, the 99 KDD Cup, and text categorization, will be presented.
Keynote Lecture 2
Impact of the ICT on the Management and Performance of Intelligent Transportation Systems
  Maria P. Fanti,
Polytechnic of Bari,
Brief Bio
Maria Pia Fanti is associate professor in Systems and Control Engineering and is with the Department of Electrical and Electronic Engineering of the Polytechnic of Bari (Italy). Maria Pia Fanti received the Laurea degree in Electronic Engineering from the University of Pisa (Italy), in 1983 and obtained an IBM thesis award. She was a visiting researcher at the Rensselaer Polytechnic Institute of Troy, New York, in 1999. Her research interests include discrete event systems, Petri nets, modeling and control of automated manufacturing systems, modeling and management of logistics system, supply chains and health care systems.

Prof. Fanti is Associate Editor of the following journals: IEEE TRANS. ON SYSTEMS, MAN, AND CYBERNETICS. PART A, IEEE TRANS. ON AUTOMATION SCIENCE AND ENGINEERING, The Mediterranean J. of Measurement and Control, Int. J. of Automation and Control, and Enterprise Information Systems. She is Co-Chair of the Technical committee on Discrete Event Systems for the IEEE SMC Society, Chair of the Italy Section SMC Chapter, and member of the IFAC Technical Committee on Discrete Event and Hybrid Systems. She is authors of 120+ papers. She has served in 20+ conference international program committees, she is IPC chair of 2nd IFAC Workshop on Dependable Control of Discrete Systems, Bari, Italy, 2009 and of the IEEE Workshop on Health Care Management, Venice, Italy, 2010.

Intermodal logistic systems (ILS) modelling, planning, and control are research streams that, in the last years, have received a significant attention by the researcher and practitioner communities due not only to their economic impact, but also to the complexity of decisional, organizational, and management problems. Nevertheless, the increasing complexity of these systems and the availability of the modern ICT (Information and Communication Technologies) for the interaction among the different decision makers and for the acquisition of information by the decision makers, require both the development of suitable models and the solution of new decision problems.

This talk is aimed at presenting the new attractive researches and projects in the field of ILS operational control and management in Europe. In particular, the talk points out the key solution of using effectively and efficiently the latest developments of ICT for ILS operational management. Moreover, it is shown how suitable models can describe the structure and the dynamics of the LIS at the real-time operational level in order to evaluate the impact of the ICT on costs and performances. To this aim, some real case studies will be presented and analyzed in different dynamic conditions characterized by a different level of information that is shared between infrastructures and operators. Moreover, the model and the simulations are employed to compare the impact of investing in transportation system infrastructures with respect to the investment in ICT.
Keynote Lecture 3
Recent Advances in Verification and Analysis of Hybrid Systems
  Janan Zaytoon,
University of Reims Champagne Ardennes,
Brief Bio
Born in 1962, Janan Zaytoon (BSc Eng./1983, MSc Eng./1986, DEA/1988, PhD/1993, Habilitation/1997) is Professor and Head of the CReSTIC Research Centre (involving 150 researchers) at Reims University. He was a member of the Administration Council of the same University (2003-2006). He is the Chair of the French national research network/group “GDR MACS of CNRS”, which involves all the researchers in the field of Automatic Control Syetems in France (about 2000 researchers and PhD students).

His involvement in IFAC includes his service as member of the IFAC Council from 2008 to 2011, head of the French National Member Organizer since 1999, Chair of Technical Committee on Discrete Event and Hybrid Systems from 2005 to 2008, Vice-Chair of this Technical Committee from 2002 to 2005 and 2008 to 2011, member of the Publication Committee of IFAC from 2008 to 2011, Editor of the IFAC Journal “Control Engineering Practice” and the Affiliated IFAC Journal “Nonlinear Analysis: Hybrid Systems”.

Professor Janan Zaytoon is the author/co-author of 70 journal papers, 3 books, 12 book chapters, 120 conference papers, and 8 patents. His main research interests are in the fields of Discrete Event Systems, Hybrid Dynamic Systems, Intelligent Control and Biomedical Engineering. He is an associate Editor of “IET Control Theory and Applications” and “Discrete Event Dynamic Systems”, IPC and/or NOC Chair/Co-Chair of 15 Conferences, Editor/Co-Editor of 10 Conference Proceedings, Keynote speaker for 6 conferences, supervisor of 20 PhD students, Guest Editor/Co-editor for 18 special issues of 6 international and 2 national journals, leader of 8 industrial contracts, and was Chair of the WODES (International Workshop on Discrete Event Systems) steering Committee.

Formal verification of properties is a very important area of analysis of hybrid systems. It is, indeed, essential to use methods and tools to guarantee that the global behaviour of a system is correct and consistent with the specifications. This is especially true for safety properties that insure that the system is not dangerous for itself or its environment.

Classically, verification of Safety properties may be performed with reachability computation in the hybrid state space. Basic ideas have not really evolved since the first works, however new techniques have been proposed and algorithms have been improved.

The aim of this talk is to present the problem of verification and reachability computation for hybrid systems and to propose a classification of recent improvements. To overcome the difficulties in verification and reachability analysis it is necessary to make choices regarding general principles, algorithms and mathematical representation of regions of the continuous state space. These choices depend on each other and must be consistent. However all approaches are based on common considerations that will be used to structure the talk.
Keynote Lecture 4
Discrete Event Diagnosis using Petri Nets
  Alessandro Giua,
Universita' di Cagliari,
Brief Bio
Alessandro Giua is professor of Automatic Control at the Department of Electrical and Electronic Engineering of the University of Cagliari, Italy. He received the Laurea degree in electric engineering from the University of Cagliari, Italy in 1988, and the M.S. and Ph.D. degrees in computer and systems engineering from Rensselaer Polytechnic Institute, Troy, New York, in 1990 and 1992.

His research interests include discrete event systems, hybrid systems, networked control systems, automated manufacturing, Petri nets, control of mechanical systems, failure diagnosis. He has co-authored two textbooks on Automatic Control (in Italian) and over 150 technical papers.

Dr. Giua is a member of the editorial board of the journals: Discrete Event Dynamic Systems: Theory and Applications; IEEE Trans. on Control Systems Technology; Nonlinear Analysis: Hybrid Systems. He has served in the program committee of over 60 international conferences.

He is chair for Chapter Activities of the Member Activities Board of the IEEE Control Systems Society and chair of the IFAC Technical Committee 1.3 on Discrete Event and Hybrid Systems.

The diagnosis of discrete event systems is a research area that has received a lot of attention in the last years and has been motivated by the practical need of ensuring the correct and safe functioning of large complex systems. Petri net models have often been used in this context: the intrinsically distributed nature of Petri nets where the notion of state (i.e., marking) and action (i.e., transition) is local has often been an asset to reduce the computational complexity involved in solving a diagnosis problem.

In this talk I will present an approach for the failure diagnosis of discrete event systems modeled by place/transition nets. The transitions of a net can be partitioned into two sets: observable or unobservable transitions (the latter set including all those transitions that model faulty behaviors). The proposed diagnosis approach is based on the notion of basis marking and justification, that allow one to characterize the set of markings that are consistent with the actual observation, and to reconstruct the sequence of unobservable transitions whose firing is required to reach them. This approach applies to all net systems whose unobservable subnet is acyclic. If the net system is also bounded the proposed approach may be significantly simplified moving the most burdensome part of the procedure off-line.
Keynote Lecture 5
Meeting the World Challenges: From Philosophy to Information Technology to Applications
  Peter S. Sapaty,
Institute of Mathematical Machines and Systems, National Academy of Sciences
Brief Bio
Dr Peter Simon Sapaty, chief research scientist and director of distributed simulation and control project at the Institute of Mathematical Machines and Systems, National Academy of Sciences of Ukraine, is with networked systems for more than four decades. A power network engineer on education, he created citywide heterogeneous computer networks from the end of the sixties, implemented a multiprocessor macro-pipeline supercomputer in the seventies-eighties, and since then used distributed computer networks for solving complex problems of most different natures—from distributed knowledge bases to intelligent network management to road traffic control to simulation of battlefields. He also worked in Germany, UK, Canada, and Japan as Alexander von Humboldt Foundation fellow, project leader, research professor, department head, and special invited professor; created and chaired a SIG on mobile cooperative technologies within Distributed Interactive Simulation project in the US. Peter invented a higher-level distributed networking technology used in different countries and resulted in a European Patent and two John Wiley books. His interests include coordination and simulation of large distributed dynamic systems under the holistic and gestalt principles, with application in advanced command and control, cooperative robotics, infrastructure protection, crisis management, and especially for finding asymmetric solutions in unpredictable and hostile environments.

We have been witnessing numerous world crises and disasters—from ecological to military to economic, with global world dynamics likely to be increasing this century further. To withstand the unwanted events and their consequences (ideally: predict and prevent them) we need effective worldwide integration of numerous efforts and often dissimilar and scattered resources and systems. Just establishing advanced communications between parts of the distributed systems and providing the possibility of sharing local and global information from any point, often called interoperability, is becoming insufficient (and even dangerous) for solving urgent problems in real time and especially ahead of it. We may need the whole distributed system to behave as an integral organism, with parts not so interoperating with each other but rather representing together an integral whole pursuing global goals and having a sort of global consciousness, with the whole being much more than the sum of the parts, the latter having sense in the context of this whole rather than vice versa. The lecture will highlight known holistic and gestalt principles mainly used in psychology and psychiatry in relation to a single brain, extending them to any distributed systems as a whole, which may need highest integrity and performance for the reaction on unpredictable situations. A higher than traditional interoperability organizational layer will be proposed that will enable any distributed resources and systems to behave as an integral organism operating under global goals. This “over-operability” layer can be established by implanting into key system points of the same copy of a universal intelligent module U, which can communicate with other such modules and interpret collectively global mission scenarios presented in a special Distributed Scenario Language (DSL). The scenarios can be injected into a distributed system from any U and then self-replicate, self-modify and self-spread throughout the system to be managed, tasking individual components, activating distributed resources, and establishing runtime infrastructures between the elements that provide integrity and “consciousness” of the system as a whole. The approach has been prototyped on different platforms and practically tested on numerous applications. The lecture will provide programming examples in DSL of exemplary problems from: distributed simulation of battlefields; cooperative robotics with efficient combination of hierarchical control with loose swarming and flexible situation-dependent watershed between the two; electronic warfare with battlespace dominance; intelligent tactical communications with dramatic reduction of (most vulnerable) central control facilities; distributed avionics where high integrity of the distributed control system could save lives even in a physically disintegrating object; directed energy systems enabling us to transfer power into any point of the globe via a network of space mirrors; terrorism and piracy fight where asymmetric solutions to asymmetric situations and threats can be programmed on the fly. The structure of the distributed DSL interpreter and details of its software or hardware implementation will be discussed too. -- This work has been sponsored by the Alexander von Humboldt Foundation in Germany.
Page updated on 7/07/09
Copyright © INSTICC