Computational Intelligence and Applications|计算智能与应用

Distinguished and Invited Speakers

Speaker I

IEEE Fellow, Prof. Robert Kozma, University of Massachusetts Amherst, MA, USA & University of Memphis, TN, USA

Speech Title: Phase Transitions in Brains: The Cognitive "Aha" Moment and Applications in Engineering Systems

Abstract: Advanced brain imaging techniques, including EEG, ECoG, fMRI, and MEG, indicate discontinuities in brain dynamics at theta rates (4-8 Hz). The observed neural processes can be interpreted as neural correlates of cognition. In particular, sudden synchronization-desynchronization transitions in brain dynamics have been identified as neural markers of higher cognition, decision making, and the so-called “aha” moment of sudden insight and understanding. 

We employ graph theoretical tools to interpret the experimental findings. Graph theoretical approaches have been extremely useful in the past decades to describe structural and functional properties on large-scale networks, including the world-wide-web, social networks, ecological networks, biological systems. Our focus here is on neural systems, in particular on brain networks. These studies lead to breakthrough of our understanding of cortical networks and dynamics. By developing random graph theory and percolation models, we provide a solid theoretical foundation of sudden changes in brains, and interpret them as cortical phase transitions. This leads to several crucial hypotheses, such as the presence of effects described as "Black Swan" and "Dragon King," and the predictability of the transient dynamics. Applications include not only cognitive engineering, but issues related to emergency response to catastrophic events, as well as transient dynamics in variable stars and supernovae formation.

 

Biography: Kozma received his MS in Power Engineering from the Moscow Power Engineering Institute in 1982, his MS in Mathematics from the Eötvös Loránd University in 1988, and his PhD in Applied Physics from Delft University of Technology in 1992.

Kozma joined the Computer Science Department of the University of Massachusetts Amherst as Visiting Professor in 2016. He is Professor of Mathematics at the University of Memphis, where he has been Associate and Full Professor of Computer Science since he joined the ranks of faculty in 2000. Previous affiliations include Department of EECS at the University of California at Berkeley (1998-2000), Lecturer at Otago University (1996-1998), Associate Professor at Tohoku University, Japan (1993-1996), Research Assistant/Fellow of the Central Research Institute of Physics of the Hungarian Academy of Sciences (1982-1988). He has held visiting positions at NASA/JPL, Sarnoff Co., Princeton, NJ; LBL, Berkeley, AFRL Hanscom AFB, MA and WPAFB, Dayton, OH. 

Professor Kozma is Fellow of IEEE, Fellow of International Neural Network Society (INNS). He is President-Elect of INNS (2016), recipient of INNS Gabor Award (2011), and has been NRC Senior Fellow (2006-2008). He is on the Governing Board of IEEE SMC Society (2016-2018); has been AdCom Member, IEEE CIS (2009-2012), Chair, Distinguished Lecturers Program, IEEE CIS (2010), General Chair, Intl. Joint Conference on Neural Networks, IJCNN09, Atlanta, GA (2009). He has been Associate Editor of IEEE Transactions on Neural Networks, Cognitive Neurodynamics, Neural Networks, Neurocomputing, Soft Computing, and other journals. He is Associate Editor of ‘Neural Networks (Elsevier),’ ‘IEEE Transactions on Neural Networks,’ ‘Neurocomputing’ (Elsevier), ‘Journal of Cognitive Neurodynamics’ (Springer), Area Editor of ‘New Mathematics and Natural Computation’ (World Scientific), and ‘Cognitive Systems Research.’ He is author of over 300 referred papers, 7 book volumes.

Speaker II

Prof. William W. Song, Dalarna University, Sweden

Speech Title: Granularity and Semantics – Macro versus Micro Data Analysis

Abstract: How to determine an appropriate view of data analysis processes and data presentations is an extremely important issue when people want to understand trivial data or common-sense concepts as results of intensive data analyses. This issue involves a precise granularity identification as well as a flexible shifting between the different layers of views, fitting various settings and satisfying different users demands. The data analysis results of a transient change analysis of traffic flows of all crosses in a large city within a minute at a morning peak time might be useless to a car driver trapped in a traffic congestion when going to work. Likewise, it does not help very much to this driver either if a general city congestion index for a day is provided to her. However, this situation happens from time to time when researchers apply data analysis methods such as Predictive Analytics (PA) or Artificial Neuro-network (ANN) but do not realize an appropriate granularity of semantic views should be considered and figured out, resulting in difficulty for the users to understand and use the big data analysis outcomes. In this talk the author intends to point out what various data analysis methods are suitable for which views in which an appropriate granularity is defined. The author like to emphasize that, for the last few years, lack of deep analysis of semantic analysis and conceptual modelling in the research field of big data analysis for the areas such as e-commerce, social network systems, public traffic systems, and general healthcare, where huge amount of data accumulated in volume along the time is a major cause that results in this granularity problem.​

Biography: William Wei Song received his BSc in computer science from Zhejiang University, Hangzhou, China in 1982 and PhD in information systems and sciences from Stockholm University and the Royal institute of Technology, Stockholm, in Sweden in 1995.

He started his career at a university in 1982. After receiving his PhD degree, he became staff researcher at SISU, Sweden from 1995, senior researcher at ETI, Hong Kong University, China from 1999, and associate professor at Durham University, UK, from 2003. He is now a full professor in Business Intelligence and Information Systems at Dalarna University, Sweden. His research interest covers a wide range of fields, including computer science, information systems, artificial intelligence, semantic web, service science, business intelligence, e-business, e-learning, and online education. He has published more than 100 research papers in international journals and conferences.

Professor Song is also guest professor (researcher) of a number of overseas universities and sits at the board of a number of international journals.

Speaker III

Prof. Manuel Núñez, Complutense University of Madrid, Spain

Speech Title: An overwiev of the use of formal methods in IMDs

Abstract: The evolution of software is the main reason of many of the failures in some systems. The relevance of these failures depends on the purpose of the systems. In the case of Implantable Medical Devices (IMDs), it is essential to ensure their security and their ability to recognize some patterns of illnesses.

Formal methods can be used to appropriately check that the expected requirements are fulfilled. Formal methods allow us to formalize the requirements and specification of the system that we want to develop by taking into account aspects such as the security of the communications with external elements and their actuation mode in case of alarm or warning according to the patient state. For example, if we develop a defibrillator then we need to ensure that it reacts in case of a cardiac arrest or if the number of beats abruptly changes.

This talk will show the importance of formal methods in this area and the current state of its application in IMDs. Some of the techniques that will be explained are timed automaton, hybrid automaton and extended finite machines and the focus will be on the three main IMDs: pacemakers, defibrillators and infusion pumps.

Biography: Manuel Núñez is a Professor in the Department of Computer Systems and Computation of the Complutense University of Madrid, Spain. He holds a Doctorate degree in Mathematics & Computer Science, obtained in 1996. Additionally, he holds a Master degree in Economics, obtained in 2002.

 

He has done research in the broad field of formal methods. Currently, he is interested in the study of formal methods for testing complex systems. Specifically, he has three main lines of research:

* Formal analysis of systems with distributed testers, in particular, those where time and probabilities play an important role.

* Passive testing of multi-user systems with asynchronous communications.

* Specification and testing of health related systems.

 

Manuel Núñez belongs to the following scientific committees:

* IEEE SMC Technical Committee on Computational Collective Intelligence,

* Board of Directors of the Tarot Summer School on Software Testing,

* ICCCI Steering Committee,

* A-MOST Workshop Steering Committee

He is a member of several Editorial Boards of journals and has served in more than 130 Program Committees of international events in Computer Science. He has published more than 130 papers in international scientific journals and meetings.

 

Speaker IV

Prof. Lipo Wang, Nanyang Technological University, Singapore

Speech Title: Natural Computation for Optimization

Abstract: This talk highlights some of our research results in optimization using intelligent techniques inspired from nature. The techniques that we use include our noisy chaotic neural network, ant colony optimization, and genetic algorithms. In particular, neural networks are intrinsically parallel systems with potential for fast hardware implementation. We demonstrate our algorithms in several challenging optimization problems, such as optimal channel assignment in mobile communications, optimal multicast routing, topological optimization problem (TOP) in backbone network design, the broadcast scheduling problem (BSP) in packet radio networks, frequency assignment problem (FAP) in satellite communications, compact radial-basis-function (RBF) neural networks, class-dependent and class-independent feature selection, neural network tuning, and image segmentation.

Biography: Dr. Lipo Wang received the Bachelor degree from National University of Defense Technology (China) and PhD from Louisiana State University (USA). His research interest is intelligent techniques with applications to optimization, communications, image/video processing, biomedical engineering, and data mining. He is (co-)author of 300 papers, of which 100 are in journals. He holds a U.S. patent in neural networks and a Chinese patent in VLSI. He has co-authored 2 monographs and (co-)edited 15 books. He was/will be keynote/panel speaker for 30 international conferences. He is/was Associate Editor/Editorial Board Member of 30 international journals, including 3 IEEE Transactions, and guest editor for 10 journal special issues. He was a member of the Board of Governors of the International Neural Network Society (for 2 terms), IEEE Computational Intelligence Society (CIS, for 2 terms), and the IEEE Biometrics Council. He served as CIS Vice President for Technical Activities and Chair of Emergent Technologies Technical Committee, as well as Chair of Education Committee of the IEEE Engineering in Medicine and Biology Society (EMBS). He was President of the Asia-Pacific Neural Network Assembly (APNNA) and received the APNNA Excellent Service Award. He was founding Chair of both the EMBS Singapore Chapter and CIS Singapore Chapter. He serves/served as chair/committee members of over 200 international conferences.

Speaker V

Prof. Ruhul Sarker, University of New South Wales, Australia

Speech Title: Adaptive Configurations of Evolutionary Algorithms

Abstract: Over the last few decades, Evolutionary Algorithms (EAs) have shown tremendous success in solving many complex optimization problems. The success of any EAs is dependent on the choice of its search operators as well as parameters. Many studies have been conducted to select the appropriate operators and parameters for EAs but with very little success. An algorithm, with a given set of operators and parameters, may work well for a problem or a set of problems, but that may perform badly with other problems. To ensure good performance of EAs over a wide range of problems, it requires a new strategy for configuring EAs. In this talk, different configurations of EAs will be discussed and their performance will be analysed by solving a wide range of test and practical constrained optimization problems.

Bio: Ruhul A Sarker obtained his Ph.D. from Dalhousie University (former TUNS), Canada. He is a Professor in the School of Engineering and IT, and the Director of Faculty PG Research at UNSW Canbarra (located at ADFA), Australia. Prof. Sarker’s broad research interests are decision analytics, CI / evolutionary computation, operations research, and applied optimization. 

Prof. Sarker is an editor of the Journal of Flexible Service and Manufacturing, an associate editor of the Journal of Memetic Computing, an editor of the Journal of Industrial and management Optimization, and former editor-in-chief of ASOR Bulletin. He had led the technical committee for IEEE Congress on Evolutionary Computation in 2003 and was a proceeding co-chair of IEEE WCCI'2012. 

Prof. Sarker is the lead author of the book ‘Optimization Modelling: A Practical Approach’ published by Taylor & Francis /CRC Press. He has edited /co-edited 8 books, on specialised topics in his areas of research, published by the leading publishers. He has 250+ refereed publications including 100+ journal papers. His research has received a number of international media coverages. Some of these media are: Wall Street JournalScienceDailyUnited Press International (UPI), Times of the Internet, TerraDaily,MarketWatchEcoEarthEuroceanSoftpediaOneindiaScienceBlog. For publication detals, visit:https://research.unsw.edu.au/people/professor-ruhul-sarker/publications

Speaker VI

Prof. Li Zhang, University of Cincinnatii, USA

Speech Title: Model of DNA breakpoints in cancer depended on genomic DNA sequence features and replication timing

Abstract: DNA breakpoints mark the endpoints of copy number variants (CNVs) in cancer genomes, which are important in carcinogenesis and development. We have analyzed the distribution of the DNA breakpoints that underlie the genomic changes in over 10,000 patients of 33 cancer types collected in The Cancer Genome Atlas (TCGA) project. Cancer genomes typically contain numerous somatic CNVs. We developed a linear regression model that predicts the distribution of DNA breakpoints observed in cancer in terms of germline DNA breakage frequency, replication time, and a number of genomic sequence features such as the density of alu units.  We found that copy number gains in tumor tend to happen in the genomic regions that are replicated early and copy number losses tend to happen in late replication regions. This pattern suggests that the background rate of CNVs may be generated from random disruption of replications, which would lead to a surplus of copy number in the early-replication regions and a deficit in the late-replication regions. We demonstrate that our model provides a framework to distinguish the cancer drivers from passengers involving CNVs.

Biography: Dr. Li Zhang received his MS in Biophysics, Tsinghua University, Beijing, China in 1988, his PhD in Computational Biology, University of North Carolina at Chapel Hill, NC, USA in 1995 and finished his Postdoctoral training in the Scripps Research Institute, San Diego, CA, USA, in 1998.

Dr. Li Zhang joined the Department of Department of Environmental Health, University of Cincinnatii as associate professor in 2016. Previous affiliations include senior scientist, Ernest Gallo Clinics and Research Center, University of California at San Francisco, CA, (1998-2001), assistant professor, Department of Biostatistics, Division of Quantitative Sciences, The University of Texas M. D. Anderson Cancer Center, Houston, TX, (2001-2006), assistant professor, Department of Bioinformatics and Computational Biology, Division of Quantitative Sciences, The University of Texas MD Anderson Cancer Center, Houston, TX, (2006-2010), associate professor, Department of Bioinformatics and Computational Biology, The University of Texas MD Anderson Cancer Center, Houston, TX, (2010-2016) and associate professor (adjunct), Department of Statistics, Rice University, Houston, Texas,  (2010-2016).

Dr. Li Zhang is one of the members of American Association for the Advancement of Science (AAAS) and International Society of Computational Biology (ISCB). 

Dr. Li Zhang is also an associate editor of International Epidemiology and Genetics, Cancer Medicine and Journal of Computational Systems Biology and he is a reviewer for many journals, such as Nature Biotechnology, PNAS, Nucleic Acid Research, Cancer Research, Bioinformatics, BMC Bioinformatics, BMC Genomics and Cancer Medicine, etc. Moreover, He has published more than 80 research papers and holds 2 U.S. patents in University of Texas M.D. Anderson Cancer Center.

Speaker VII

Prof. Rafał Scherer, Institute of Computional Intelligence, Poland

Speech Title: Convolutional Neural Networks – From Images to Other Types of Data

Abstract: Inspired by biological processes, convolutional neural networks (CNN) proved to be successful in classifying large, diverse, multi-class image datasets.  This come from multiple improvements over conventional multilayer neural networks. Convolution operation sliding over the image with shared weights reduces the number of trainable parameters, improves generalizations and makes CNN immune partially to various input image transformations. Neurons are organized in three-dimensional structures to cope with the third dimension of the input data. Pooling layers allows to gradually reduce the spatial size of the features. By rearranging input data, CNN can deal effectively with data types other than images, e.g. letters, texts or numerical and textual streams such as network traffic.

Biography: Rafał Scherer is an associate professor at the Częstochowa University of Technology and head of Computer Vision and Data Mining Lab. His research focuses on developing new methods in computational intelligence and data mining, ensembling methods in machine learning, content-based image indexing and classification. He authored more than 80 research papers and a book on multiple classification techniques published in Springer. He was a reviewer for major computational intelligence journals. Scherer earned MSc degree in electrical engineering at Department of Electrical Engineering and PhD degree in computer science (Methods of Classification Using Neuro-Fuzzy Systems) at the Department of Mechanical Engineering and Computer Science of Czestochowa University of Technology. He co-organizes every year or two years the International Conference on Artificial Intelligence and Soft Computing in Zakopane (http://www.icaisc.eu/) which is one of the major events on computational intelligence. He is also a co-editor of the Journal of Artificial Intelligence and Soft Computing Research (http://jaiscr.eu/).