Invited Speakers
Prof. Michael Pecht, University of Maryland, USA
Prof. Yukio Ohsawa, University of Tokyo Japan
Prof. Sabri Pllana, Linnaeus University, Sweden
Prof. Karim Djouani, University Paris Est Creteil (UPEC), Paris, France
Prof. Mourad Fakhfakh, National School of Electronics and Telecommunications of Sfax, Tunisia
Prof. Kaushik Das Sharma, University of Calcutta, India.
Prof. Ali Siadat, Ecole Nationale Sup rieure d'Arts et M tiers (ENSAM), France
Prof. Fabio Scotti, Università degli Studi di Milano, Italy
S1: Title: Addressing Hardware Security in Embedded Systems
| Prof Michael Pecht, Director of Center for Advanced Life Cycle Engineering University of Maryland, USA |
Abstract: With the latest developments in networking technologies, software attacks are becoming more frequent across a wide range of industries. However, not all attacks on a system are software caused or related; the physical components of a system can also be compromised and in some cases may create anomalies that could appear as a cyber-attack. This is especially a concern because in most systems today, components, manufacturing, and assembly are often outsourced to offshore facilities. The result is that anomalous behavior can be due to defective components, counterfeit components, or Trojan horses. We propose a deep learning based method to detect anomalous system behavior and classify it to natural aging (degrading components), or ?maliciously induced? aging i.e. counterfeit or hardware attacks.
Biography: Prof Michael Pecht is a world renowned expert in strategic planning, design, test, and risk assessment of electronics and information systems. Prof Pecht has a BS in Physics, an MS in Electrical Engineering and an MS and PhD in Engineering Mechanics from the University of Wisconsin at Madison. He is a Professional Engineer, an IEEE Fellow, an ASME Fellow, an SAE Fellow and an IMAPS Fellow. He is the editor-in-chief of IEEE Access, and served as chief editor of the IEEE Transactions on Reliability for nine years, and chief editor for Microelectronics Reliability for sixteen years. He has also served on three U.S. National Academy of Science studies, two US Congressional investigations in automotive safety, and as an expert to the U.S. Food and Drug Administration (FDA). He is the founder and Director of CALCE (Center for Advanced Life Cycle Engineering) at the University of Maryland, which is funded by over 150 of the world's leading electronics companies at more than US$6M/year. The CALCE Center received the NSF Innovation Award in 2009 and the National Defense Industries Association Award. Prof Pecht is currently a Chair Professor in Mechanical Engineering and a Professor in Applied Mathematics, Statistics and Scientific Computation at the University of Maryland. He has written more than twenty books on product reliability, development, use and supply chain management. He has also written a series of books of the electronics industry in China, Korea, Japan and India. He has written over 700 technical articles and has 8 patents. In 2015 he was awarded the IEEE Components, Packaging, and Manufacturing Award for visionary leadership in the development of physics-of-failure-based and prognostics-based approaches to electronic packaging reliability. He was also awarded the Chinese Academy of Sciences President's International Fellowship. In 2013, he was awarded the University of Wisconsin-Madison's College of Engineering Distinguished Achievement Award. In 2011, he received the University of Maryland's Innovation Award for his new concepts in risk management. In 2010, he received the IEEE Exceptional Technical Achievement Award for his innovations in the area of prognostics and systems health management. In 2008, he was awarded the highest reliability honor, the IEEE Reliability Society's Lifetime Achievement Award. He has previously received the European Micro and Nano-Reliability Award for outstanding contributions to reliability research, 3M Research Award for electronics reliability analysis, and the IMAPS William D. Ashman Memorial Achievement Award for his contributions in reliability assessment methods for electronics products and systems. Professor Michael Pecht's research focuses on prognostics and systems health management (PHM) using machine learning. PHM is an approach that is used to evaluate the reliability of a system in its actual life-cycle conditions, determine the initiation of failure, and mitigate system risks. Prognostics of a system can yield an advance warning of impending failure in a system and thereby help in maintenance and corrective actions.. The outputs of a prognostic assessment of a product are the failure risk, time to failure, remaining useful life, and a prognostic distance within which time specific maintenance and repair actions can be taken to extend the life of the product.. The U.S. Joint Strike Fighter (JSF) Program requires PHM. NASA uses the Integrated Vehicle Health Management (IVHM) program for its fleet. Consumer electronics companies, including computer companies such as Dell, are investing a lot of money in prognostics research so that they can harness the benefits of PHM for reducing warranty costs and cutting product qualification time. The data-driven and fusion approaches stand among the three main approaches to implementing prognostics for a system (along with model-based). The data-driven prognostics methods use current and historical data to statistically and probabilistically derive decisions, estimates, and predictions about the health and reliability of products. Data-driven approaches are useful to monitor the health of large multivariate systems and are capable of intelligently detecting and assessing correlated trends in the system dynamics to estimate the current and future health of the system. Areas of interest for data-driven approaches include anomaly detection, fault identification, fault isolation and prediction of remaining useful life (prognostics). Machine learning is highly used in the data-driven approach since it incorporates statistical and probability theory in addition to data preprocessing, dimensionality reduction by compression and transformations, feature extraction, and cleaning (de-noising) of data. Fusion methods for prognostics offer the benefits of model-based and data-driven methods.
S2: Title: Data Jackets as Communicable Metadata for Potential Innovators
| Prof. Yukio Ohsawa, School of Engineering University of Tokyo Japan |
Abstract:Data Jackets are human-made metadata for each dataset,
reflecting peoples' subjective or potential interests. Via
visualizing relevances between DJs, participants in the market of data
think and talk about why and how they should combine the corresponding
datasets. Even if the owners of data may hesitate to open their data
to the public, they can present the DJs in the data marketplace that
can be regarded as a platform for data-driven innovations. In this
market, the participants communicate to find ideas to combine/use data
and find future collaborators. Furthermore, explicitly or implicitly
required data can be searched by use of tools such as DJ Store and
Variable Quest, created on DJs, which enabled analogical inventions of
data analysis methods. As a result, the innovators' marketplace on DJs
turned out to be the birthplace of fruits in business and science.
Biography: Yukio Ohsawa is a professor of Systems Innovation in the School of Engineering, The University of Tokyo. He received BE, ME, and PhD in from the School of Engineering, The University of Tokyo (1995). Then worked for the School of Engineering Science in Osaka University (research associate, 1995-1999), Graduate School of Business Sciences in University of Tsukuba (associate professor, 1999-2005), and moved back to The Univ. of Tokyo. He started researches from non-linear optics, and, via artificial intelligence, created a new domain chance discovery meaning to discover events of significant impact on decision making, since year 2000. About chance discovery he gave keynote talks in conferences such as International Symposium on Knowledge and Systems Sciences, Int?l Conf. on Rough Sets and Fuzzy Sets, Joint Conf. on Information Sciences, Knowledge-Based Intelligent Information and Engineering Systems, etc. Chance discovery came to be embodied as innovators? marketplace, a methodology for innovation borrowing principles of the dynamics of markets. Then he, when biking from his job in a business school, invented the basic idea of Data Jackets. Since then, he is introducing the method presented in this book to sciences, educations, and businesses. His original concepts and technologies have been published as books and monographs from global publishers such as Springer Verlag, Taylor & Francis, etc. Two most important books among them are, ?Chance Discovery? (2003 Springer, foreword given by Eric von Hippel), ?Innovators? Marketplace: Using Games to Activate and Train Innovators? (2012 Springer, foreword given by Larry Leifer). He edited special issues as guest editors for journals, mainly relevant to chance discovery, such as Intelligent Decision Technologies (2016), Information Sciences (2009), New Generation Computing (2003), Journal of Contingencies and Crisis Management (2002), etc.?As the program chair of the Annual Conference of The Japanese Society on Artificial Intelligence, he came to be the first to change this domestic conference into an international conference from June 2019.
S3: Title: Meta-heuristics and Machine Learning Methods for Optimization of Software for Parallel Computing Systems
| Prof. Sabri Pllana Linnaeus University Department of Computer Science and Media Technology Sweden |
Abstract:Optimal software execution on parallel computing systems demands consideration of many parameters at compile-time and run-time. Determining an optimal set of parameters in a given execution context is a complex task, and therefore to address this issue researchers have proposed different approaches that use meta-heuristics or machine learning.
In this talk, we review approaches that use machine learning or meta-heuristics for software optimization at compile-time and run-time. Thereafter, we describe our approach that combines meta-heuristics and machine learning for optimal workload distribution on heterogeneous computing systems. We conclude with an overview of challenges and future research directions.
Biography: Sabri Pllana is an Associate Professor at the Department of Computer Science of the Linnaeus University in Sweden. Before joining the Linnaeus University, he worked for 12 years at the Research Group Scientific Computing of the University of Vienna in Austria. He holds a PhD degree (with distinction) in computer science from the Vienna University of Technology. Sabri Pllana is head of the High-performance Computing Centre at LNU. His research interests include intelligent methods for system optimization, cognitive computing techniques, and heterogeneous computing systems. He also studies properties of various kinds of socio-technical systems (such as, health-care systems or manufacturing systems) using innovative modelling and simulation techniques. He contributed to several EU-funded projects and he coordinated the FP7 project PEPPHER. Sabri Pllana is Senior Member of the IEEE (# 80432603), associate editor of Computing journal (Springer), member of the European Network on High Performance and Embedded Architecture and Compilation (HiPEAC), associate member of ETP4HPC, and member of the Euro-Par Advisory Board.
S4: Title: Machine learning for intention and behaviour analysis in the context of assistive mobility
 |
Prof. Karim Djouani University Paris Est Creteil (UPEC), Paris, France SARChI Chair - Tshwane University of Technology (TUT), Pretoria, South Africa |
Abstract:Assistive technologies and smart environments aim to assist users with specific needs in their daily life activities. The related systems based their decision on the analysis of their interaction with the user in order to provide both physical and cognitive assistance. Information is gathered from different sensing capabilities on the user intention and behaviour, the user and the environment contexts and the available systems and services, as the basis for the decision-making systems on how and when to deliver the right needed assistance in the right time.
Among the set of possible complex interaction, the present talk covers the aspect related to mobility aid for people with disabilities.
The user intention detection is of main importance along with the assistive control strategy of a mechatronic system, such as electrical wheelchair or wearable robots (exoskeletons) with the human in the loop.
The talk will cover hybrid brain computer (BCI) systems, used to gather EEG signals in addition to other signals for user’s intention detection as well as advanced machine learning techniques and robust control strategies with the human in the loop.
More detailed discussion will cover theoretical aspects of pattern recognition and machine learning frameworks for nonlinear, non-Gaussian signals by reviewing the state of art and the recent trends of non-linear machine learning and robust pattern recognition methods.
Biography:Karim Djouani is a Scientist and a Technical Group Supervisor of pattern recognition, machine learning, soft computing, networking systems, and robotics. He is full Professor at University Paris Est Creteil (UPEC). Since January 2014, he is the recipient DST/NRF SARChI Chair in Enabled Environment for Assistive Living at the Tshwane University of Technology (TUT), Pretoria, South Africa. In January 2011, he was appointed as a Full Professor at French South African Institute of Technology (F’SATI) at TUT. From July 2008 to December 2010, he was seconded by the French Ministry of Higher Education to F’SATI at TUT. Until July 2008, he was also Manager of national and European projects at the LISSI Laboratory. His research interests include development of novel and highly efficient algorithms for reasoning systems with uncertainty as well as optimisation, for distributed systems, networked control systems, wireless ad-hoc network, wireless and mobile communication, and wireless sensors networks as well as robotics.
He authored/co-authored more than 200 articles in archival journals, patents, conference proceedings as well as 18 chapters in edited books and 2 books.
He is a member of IEEE communication, computer, robotics and automation and Artificial Intelligence societies and several French National Research task Group (GDR-MACS, GDR-ISIS).
S5: Title: Analog Circuit Design Optimization, What a challenge!
 |
Prof. Mourad Fakhfakh Director of the ESSE research Lab. (Laboratory of Advanced Electronic Systems and Sustainable Energy) National School of Electronics and Telecommunications of Sfax, Tunisia |
Abstract:The ever growing need for higher performances of analog, mixed signal and radiofrequency (AMS\RF) circuits and systems, in addition to the current trend towards miniaturization of electronic devices and gadgets has hardened (and continues to) the design task of such appliances. Indeed, the progress of process technology has made possible manufacturing a large number of transistors on a single chip, which is very needed particularly for portable devices, but, on the other hand, it has introduced a new spectrum of problems, such as process variation, yield and reliability. These problems are making sizing/optimization process very difficult and very time consuming. As a mean of fact, nowadays, designers are more than ever focusing on developing computer aided design tools.
The proposed talk will focus on these challenges. It will put the light on main faced problems in analog and RF circuit performance optimization, highlight some proposed viable solutions, and present some circuit sizing/optimization applications.
My presentation is divided into three parts. First, I will briefly talk about main techniques used in the conventional metaheuristic - based sizing/optimization kernels, and highlight their main limitations. Then, I will look at newly proposed techniques to reduce computation time, which include rapid metaheuristics, geostatistical inspired metamodeling techniques and performance estimation through ‘Monte Carlo’ analysis.
Finally, I will put the spot on some newly proposed hybrid sizing/optimizing kernels, and I will offer some specific application examples (analog and RF circuit performance optimization).
Biography: Dr. Mourad Fakhfakh was born in Sfax-Tunisia in 1969. He received the engineering, the PhD and the Habilitation degrees from the national engineering school of Sfax Tunisia in 1996, 2006 and 2011, respectively. From 1998 to 2004 he worked for the Tunisian National Company of Electricity and Gas (STEG) as the head of the technical interventions’ department. In September 2004, he joined the National School of Electronics and Communications (ENET'Com), formerly (ISECS), where he is working as a full Professor. He is the head of the Advanced Electronic Systems and Sustainable Energy research laboratory (ESSE). Pr. Fakhfakh is an IEEE senior member. He was the IEEE CEDA Tunisia Chapter Chair (Council on Electronic Design Automation). He has co-edited seven books and more than 150 papers in many prestigious journals, international conferences and book chapters. He is member of the scientific committees of many international conferences. He has organized many international scientific events, such us the IEEE Tunisia CEDA’s ENG-OPTIM’ Contest (4 editions). He was the general chair of SMACD 2010. Pr. Fakhfakh research interests include analog circuit design automation, symbolic analysis techniques, modelling techniques, and applied optimization techniques.
S6: Title: Computational Intelligence Guided Bio-sensors for Cancer Sub group Detection
 |
Prof. Kaushik Das Sharma |
Abstract:These days, microarray gene expression data are playing an essential role in cancer classifications. However, due to the availability of small number of effective samples compared to the large number of genes in microarray data, many computational methods have failed to identify a small subset of important genes. Therefore, it is a challenging task to identify small number of disease-specific significant genes related for precise diagnosis of cancer sub classes. A stochastic optimization technique like particle swarm optimization (PSO) along with adaptive K-nearest neighborhood (KNN) based gene selection technique has been proposed to distinguish a small subset of useful genes that are sufficient for the desired classification purpose. This proposed technique of finding minimum possible meaningful set of genes is applied on benchmark microarray datasets, namely the small round blue cell tumor (SRBCT) data, the acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML) data. Results demonstrate the usefulness of the proposed method in terms of classification accuracy on blind test samples, number of informative genes and computing time. Further, the usefulness and universal characteristics of the identified genes would be employed to design a nano-bio sensor for particular cancer type detection.
Biography:Kaushik Das Sharma received the B.Tech. and M.Tech. degrees in Electrical Engineering from Department of Applied Physics, University of Calcutta, Kolkata, India, in 2001, 2004 respectively and Ph.D. degree from Jadavpur University, Kolkata, India, in 2012. He is currently serving as an Associate Professor in Electrical Engineering Section of Department of Applied Physics, University of Calcutta, Kolkata, India. He is a recipient of the Kanodia Research Scholarship in 2002 and University Gold Medal in 2004 from University of Calcutta. He was invited at the University of Paris-Est Creteil, France as Visiting Teacher-Researcher Fellow during June-July, 2019.
Dr. Das Sharma’s key research interests include fuzzy control, stochastic optimization, machine learning, robotics and computational biology. Dr. Das Sharma has authored/coauthored about 65 technical articles in reputed international journals and conferences.
Dr. Das Sharma is a Senior Member of the IEEE (USA), Member of the IET (UK) and a Life Member of Indian Science Congress Association. He has served/is serving in important positions in several International Conference Committees all over the world. Presently he serves as Secretary of IEEE Joint CSS-IMS Kolkata Chapter, an Executive Committee Members of IEEE Kolkata Section and IET (UK) Kolkata Local Network.
S7: Title: News paradigms for Production Systems design
| Prof. Dr. Ali Siadat, Ecole Nationale Sup rieure d'Arts et M tiers (ENSAM), France |
Abstract: Today's production systems are characterized by waste elimination, cycle time control and high work specifications for answering to reactivity and productivity constraints. Therefore, production systems must have the needed flexibility to better cushion this variability and its impact on/from human factors. Operators, although being integral part of system, are usually severely simplified during design and optimization of production systems. In recent years, the need for control over operator jobs in manufacturing has been identified as crucial for both improve industrial system performances and preserve operators? health. Since 7 years, LCFC laboratory of Arts et Metiers created a joint research laboratory with INRS (National Institute of Research in Safety) to study development process of new production system taking to account safety and health factors of workers and also to show impact of this consideration on productivity and industrial performances. In this presentation, we resume some of our research results based on a global framework established from FBS approach and focus on 3 axes of this project: human risk identification during design, production planning and optimization and process control based on data analysis.
Biography: Ali SIADAT received his PhD in robotics and digital signal processing from Institut National Polytechnique de Lorraine in France. He stated his carrier as associate professor in computer and industrial engineering at Arts et Metiers and became full professor in 2016 at the same engineering school. Since 2011, he is head of applied mathematics and computer Engineering department and deputy Director ofDesign, Manufacturing and Control laboratory. He served also as Guest Professor in several universities in China, Iran, Morocco and Brazil. His research interests include artificial intelligence, information system, knowledge formalization and operations research applied to manufacturing fields. He has published over than 60 papers in distinguished scientific journals and more than 80 papers in international conferences and he supervised more than 20 PhD theses. Pr. Siadat is actively involved in Factories of the Future program in France and his research activities are drive usually with international collaborations and very closed to industries (STMicroelectronics, Siemens, PSA, Plastic Omnium ?). He serves as Associate Editor of International Journal of Production Research and as Editorial Board of Smart Science journal. His new research projects interest to integrated human factors to optimization of production systems.
S8: Title: Design of Artificial Intelligence Systems in Biometric Applications
| Prof. Fabio Scotti Università degli Studi di Milano, Italy |
Abstract: The number of biometric applications and devices is continuously growing on a global scale and biometrics is pervasively entering the everyday life of users. This relevant expansion is producing new challenges and requirements to be fulfilled by the designers. Features such as adaptability, enhanced interactions with the user, robustness to non-ideal conditions, real-time capability, and high accuracy are strongly required in innovative applications and solutions such as cyber security, smart devices, and ambient intelligent infrastructures. The presentation will focus on innovative biometric recognition approaches and systems, with specific focus on recent technologies based on artificial intelligence and deep learning techniques. In traditional biometric systems, the designer needs to develop algorithms to extract a set of discriminative features from data. Artificial intelligence methods, and in particular Deep learning approaches, are capable to learn discriminative features directly from complex multidimensional signals and increase the accuracy, adaptability, and robustness to non-ideal conditions of biometric systems with respect to traditional approaches. The talk presents biometric systems from a technological point of view and provides an excursus of recent artificial intelligence approaches, including deep learning methods with current strong points and limitations. A review of new applications of biometric systems and recent trends will be also presented.
Biography: Fabio Scotti received the Ph.D. degree in computer engineering from the Politecnico di Milano, Milan, Italy, in 2003. He has been an Associate Professor in Computer Science with the Università degli Studi di Milano, Italy, since 2015. Original results have been published in over 120 papers in international journals, proceedings of international conferences, books, book chapters, and patents. His current research interests include biometric systems, machine learning and computational intelligence, signal and image processing, theory and applications of neural networks, three-dimensional reconstruction, industrial applications, intelligent measurement systems, and high-level system design. He was the Director of the Biometric Systems Laboratory of the Department of Computer Science, Università degli Studi di Milano, Italy and he is the Deputy Director of the Industrial, Environmental and Biometric Informatics Laboratory of the Department of Computer Science, Università degli Studi di Milano, Italy. He is IEEE Senior Member. Dr. Scotti is an Associate Editor of the IEEE Transactions on Human-Machine Systems and Soft Computing (Springer). He has been an Associate Editor of the IEEE Transactions on Information Forensics and Security, and a Guest Coeditor for the IEEE Transactions on Instrumentation and Measurement.