Committee on Informatics of the Polish Academy of Sciences

Invited Talks

   

"Model-free Fault Diagnosis in Sensor Networks"
Cesare Alippi
Politecnico di Milano, Italy and Università della Svizzera Italiana, Switzerland

CESARE ALIPPI received the degree in electronic engineering cum laude in 1990 and the PhD in 1995 from Politecnico di Milano, Italy. Currently, he is a Full Professor of information processing systems with the Politecnico di Milano. He has been a visiting researcher at UCL (UK), MIT (USA), ESPCI (F), CASIA (RC), USI(CH), A*STAR (SIN).
Alippi is an IEEE Fellow, Vice-President education of the IEEE Computational Intelligence Society, member of the Board of Governors of the International Neural Networks Society, Associate editor (AE) of the IEEE Computational Intelligence Magazine, past AE of the IEEE-Trans Instrumentation and Measurements, IEEE-Trans. Neural Networks, and member and chair of other IEEE committees.
In 2016 he received the INNS Gabor award and the IEEE Transactions on Neural Networks and Learning Systems outstanding paper award; in 2004 the IEEE Instrumentation and Measurement Society Young Engineer Award; in 2011 has been awarded Knight of the Order of Merit of the Italian Republic; in 2013 he received the IBM Faculty Award.
Among the others, Alippi was General chair of the International Joint Conference on Neural Networks (IJCNN) in 2012, Program chair in 2014, Co-Chair in 2011 and General chair of the IEEE Symposium Series on Computational Intelligence 2014.
Current research activity addresses adaptation and learning in non-stationary environments and Intelligent embedded systems.
Alippi holds 5 patents, has published in 2014 a monograph with Springer on “Intelligence for embedded systems” and (co)-authored about 200 papers in international journals and conference proceedings.
Home Page: http://home.dei.polimi.it/alippi/

Abstract
Availability and usability of data coming from a process/environment, e.g., those generated by a sensor network, introduce serious issues about their quality. In fact, not rarely acquired measurements are affected by sensor aging and faults which might introduce errors impacting on the correctness of the subsequent decision making process. The ability to detect faults is a mandatory step, which cannot be underestimated or neglected in real deployments.
In this direction, Fault Diagnosis Systems (FDS) are tools designed to supervise a process operation in order to detect, isolate and identify potential faults and, possibly, design accommodation actions. However, most FDS assume that some of -not necessarily amenable- hypothesis are satisfied, e.g., a description for the process is available; the system model is linear; a fault dictionary containing the fault signatures is provided; the nature of the fault profile and its development are known.
Current research in machine learning aims at removing/weakening the above assumptions so that FDS can be designed directly from available data, possibly within a cognitive framework.
The talk will focus on aspects related to the design of cognitive FDSs for sensor networks able to discriminate between faults, changes in the environment and model bias within an evolving framework.

 
   

"Internet of Things (IoT) Analytics"
Albert Bifet
Telecom ParisTech, France

Albert Bifet is Associate Professor at Telecom ParisTech and Honorary Research Associate at the WEKA Machine Learning Group at University of Waikato. Previously he worked at Huawei Noah's Ark Lab in Hong Kong, Yahoo Labs in Barcelona, University of Waikato and UPC BarcelonaTech. He is the author of a book on Adaptive Stream Mining and Pattern Learning and Mining from Evolving Data Streams. He is one of the leaders of MOA and Apache SAMOA software environments for implementing algorithms and running experiments for online learning from evolving data streams. He is serving as Co-Chair of the Industrial track of IEEE MDM 2016, ECML PKDD 2015, and as Co-Chair of BigMine (2015, 2014, 2013, 2012), and ACM SAC Data Streams Track (2016, 2015, 2014, 2013, 2012)
Home Page: http://albertbifet.com/

Abstract
Big Data and the Internet of Things (IoT) have the potential to fundamentally shift the way we interact with our surroundings. The challenge of deriving insights from the Internet of Things (IoT) has been recognized as one of the most exciting and key opportunities for both academia and industry. Advanced analysis of big data streams from sensors and devices is bound to become a key area of data mining research as the number of applications requiring such processing increases. Dealing with the evolution over time of such data streams, i.e., with concepts that drift or change completely, is one of the core issues in stream mining. In this talk, I will present an overview of data stream mining, and I will introduce some popular open source tools for data stream mining.

 
   

"Trends in Data Stream Mining"
João Gama
LIAAD INESC TEC and FEP University of Porto, Portugal

Bio: João Gama received his Ph.D. in Computer Science in 2000. He is a senior researcher at INESC TEC. He has worked in several National and European projects on Incremental and Adaptive learning systems, Ubiquitous Knowledge Discovery, Learning from Massive, and Structured Data, etc. He served as Program chair at several Machine Learning and Data Mining conferences. He is author of a monography on Knowledge Discovery from Data Streams and more than 200 peer-reviewed papers in areas related to machine learning, data mining, and data streams.

 
   

"Towards cognitive socio-economic modeling: a crucial role of human judgments, psychological biases, desires and intentions"
Janusz Kacprzyk
Fellow of IEEE, IFSA, EurAI(ECCAI), SMIA
Full Member, Polish Academy of Sciences
Member, Academia Europaea
Member, European Academy of Sciences and Arts
Foreign Member, Bulgarian Academy of Sciences
Foreign Member, Spanish Royal Academy of Economic and Financial Sciences (RACEF)
Systems Research Institute, Polish Academy of Sciences
ul. Newelska 6, 01-447 Warsaw, Poland
Email: kacprzyk@ibspan.waw.pl


Janusz Kacprzyk graduated from Warsaw University of Technology, Poland, with M.Sc. in automatic control and computer science, obtained in 1977 Ph.D. in systems analysis and in 1991 D.Sc. in computer science. He is Professor of Computer Science at the Systems Research Institute, Polish Academy of Sciences, and at WIT – Warsaw School of Information Technology, and Professor of Automatic Control at PIAP – Industrial Institute of Automation and Measurements. He is Honorary Foreign Professor at the Department of Mathematics, Yli Normal University, Xinjiang, China, and Visiting Scientist at RIKEN Brain Research Institute, Tokyo, Japan. He is Full Member of the Polish Academy of Sciences, Member of Academia Eueopaea (Informatics), Member of European Academy of Sciences and Arts (Technical Sciences), Foreign Member of the Spanish Royal Academy of Economic and Financial Sciences (RACEF), and Foreign Member of the Bulgarian Academy of Sciences. He is Fellow of IEEE, IFSA, EurAI (ECCAI) and MICAI.
He has been a frequent visiting professor in the USA, Italy, UK, Mexico, China. He has been a member of evaluation commissions of many foreign universities. His main research interests include the use of modern computation computational and artificial intelligence tools, notably fuzzy logic, in decisions, optimization, control, data analysis and data mining, with applications in databases, ICT, mobile robotics, systems modeling etc. He authored 6 books, (co)edited more than 100 volumes, (co)authored ca. 550 papers, including ca. 80 in journals indexed by the WoS. His bibliographic data are: due to Google Scholar - citations: 19491; h-index: 64, due to Scopus: citations: 5241; h-index: 32; due to WoS: citation: 4212, h-index: 27. He is the editor in chief of 6 book series at Springer, and of 2 journals, and is on the editorial boards of ca. 40 journals. He is a member of the Adcom of IEEE CIS, and was a Distinguished Lecturer of IEEE CIS.
He received many awards: 2006 IEEE CIS Pioneer Award in Fuzzy Systems, 2006 Sixth Kaufmann Prize and Gold Medal for pioneering works on soft computing in economics and management, 2007 Pioneer Award of the Silicon Valley Section of IEEE CIS for contribution in granular computing and computing in words, 2010 Award of the Polish Neural Network Society for exceptional contributions to the Polish computational intelligence community, IFSA 2013 Award for his lifetime achievements in fuzzy systems and service to the fuzzy community, and the 2014 World Automation Congress Lifetime Award for contributions in soft computing, the 2016 Award of the International Neural Network Society – Indian Chapter for Outstanding Contributions to Computational Intelligence. He is President of the Polish Operational and Systems Research Society and Past President of International Fuzzy Systems Association.

Abstract
This work concerns the problem of socio-economic modeling, more specifically, regional modeling using a multistage planning model over some planning horizon, for instance 5 – 20 years, which is formulated in terms of some finite (a few, usually) scenarios of development which is meant as both investments to be spent and some goals to be attained. The model is represented by a finite state dynamic system in which the state (output) is equated with some life quality indicators, and the input (control) is the amount of some investment and other expenditures which imply changes in the values of life quality indicators. This model was developed in its conceptual form in the late 1970s – early 1980s, during the author’s long time work for various regional development projects at the International Institute for Applied Systems Analysis (IIASA) in Laxenburg, Austria, one for the best known and most prestigious think tank in the broadly perceived systems analysis. Then, it was extended by using various fuzzy tools and techniques to model imprecision, as well as various more traditional probabilistic tools and techniques to model uncertainty, and – recently – extended by adding some nonconventional aggregation operators, affective computing, analysis of desires and intentions, cognitive informatics, etc. Over the decades, the model has been employed for various regional studies all over the world, notably the Upper Noteć in Poland, Tisza in Hungary, Kinki in Japan, The Danube Watershed, etc. The model has been mentioned as one of the best examples of fuzzy modeling in the special volume published for the 50th anniversary of the British Operational Research Society (cf. Thomas L.C., Ed. (1987) Golden Developments in Operational Research, Pergamon Press, New York, and for its fuzzy dynamic programming based algorithm the author has been awarded the 2006 Fuzzy Pioneer Award of the IEEE Computational Intelligence Society. Basically, at each planning stage, the region, presented as a system under control, with the states (outputs) equated with some life quality indicators, is subject to some investments (spending) due to some development scenario, which implies a change of the values of the life quality indicators assumed. These changes occur at each planning stage, over some assumed horizon, and their temporal evolution implies a trajectory of regional development. The basic problem is to find such investments (expenditures) which would imply the best possible evaluation of the trajectory. This evaluation is a very complicated issue because of many sophisticated, both objective and human judgment based, evaluations and/or assessments which can be measured and just perceived, but both are related to some satisfaction of the goodness of development. This is, however, a very complicated problem. First, there are two main actors, the authorities (usually national as regions may be not rich enough) who are planning and responsible for the development, and inhabitants for whom the development is to proceed. Basically, the authorities would rather be more cost conscious while the inhabitants – more effect (better values of life quality indicators) conscious. Both actors would prefer some stability in the sense of a possibly evenly distributes costs and effects over the planning horizon, and also some fair division of funds meant for the improvement of the particular life quality indicators., and a proper aggregation of satisfaction from the values of the life quality indicators attained. Moreover, all evaluations and assessment can be both objective, resulting from the sheer values of attained, and subjective, resulting from an analysis of intentions, desires, etc., and taking into account some psychological biases, notably the so called status quo bias which basically means that the human being prefer smaller changes. An important aspect is a fairness orientation of both the authorities and inhabitants, maybe with more emphasis on the so called outcome-based inequity aversion approach, in the case of authorities, and the intention-based reciprocity approach, in the case of inhabitants. In all these cases a sophisticated fusing of outcomes related to the particular life quality indicators and their temporal distribution, actors, and their specific judgmental characteristics, etc. is to be applied. We use some results of behavioral economics, psychology, intention modeling, affective computing, Wang’s cognitive informatics, etc. Moreover, we show the use some elements of "natural language technology", notably natural language generation in the context of linguistic summarization. Virtually all the above aspects of socio-economic development, and its evaluation/assessment, related to life quality indicators and distributed over the planning horizon, involves clearly both much uncertainty and imprecision. We show mainly the imprecision related aspects, dealt with using fuzzy logic, develop a multistage planning model, and show some examples.

 
   

"Spiking Neural Networks: The Machine Learning Approach"
Nikola Kasabov, FIEEE, FRSNZ (Web Page)
Knowledge Engineering and Discovery Research Institute (KEDRI), Auckland University of Technology, New Zealand

Professor Nikola Kasabov is Fellow of IEEE, Fellow of the Royal Society of New Zealand and DVF of the Royal Academy of Engineering, UK. He is the Director of the Knowledge Engineering and Discovery Research Institute (KEDRI), Auckland. He holds a Chair of Knowledge Engineering at the School of Computing and Mathematical Sciences at Auckland University of Technology. Kasabov is a Past President and Governor Board member of the International Neural Network Society (INNS) and also of the Asia Pacific Neural Network Assembly (APNNA). He is a member of several technical committees of IEEE Computational Intelligence Society and a Distinguished Lecturer of the IEEE CIS (2012-2014) He is a Co-Editor-in-Chief of the Springer journal Evolving Systems and has served as Associate Editor of Neural Networks, IEEE TrNN, IEEE TrFS, Information Science, J. Theoretical and Computational Nanosciences, Applied Soft Computing and other journals. Kasabov holds MSc and PhD from the TU Sofia, Bulgaria. His main research interests are in the areas of neural networks, intelligent information systems, soft computing, bioinformatics, neuroinformatics. He has published more than 600 publications that include 15 books, 180 journal papers, 80 book chapters, 28 patents and numerous conference papers. He has extensive academic experience at various academic and research organisations in Europe and Asia, including: TU Sofia, University of Essex, University of Otago, Advisor- Professor at the Shanghai Jiao Tong University, Guest Professor at ETH/University of Zurich. Prof. Kasabov has received the APNNA "Outstanding Achievements Award", the INNS Gabor Award for "Outstanding contributions to engineering applications of neural networks", the EU Marie Curie Fellowship, the Bayer Science Innovation Award, the APNNA Excellent Service Award, the RSNZ Science and Technology Medal, and others. He has supervised to completion 38 PhD students. More information of Prof. Kasabov can be found on the KEDRI web site: http://www.kedri.aut.ac.nz.

Abstract
The current development of the third generation of artificial neural networks - the spiking neural networks (SNN) along with the technological development of highly parallel neuromorphic hardware systems of millions of artificial spiking neurons as processing elements, makes it possible to model complex data streams in a more efficient, brain-like way [1,2]. The talk first presents some principles of deep learning implemented in a recently proposed evolving SNN (eSNN) architecture called NeuCube. NeuCube was first proposed for brain data modelling [3,4]. It was further developed as a general purpose SNN development system for the creation and testing of spatio/spectro temporal data machines (STDM) to address challenging data analysis and modelling problems. A version of the development system is available free from: http://www.kedri.aut.ac.nz/neucube/, along with papers and case study data. The talk introduces a methodology for the design and implementation of SNN systems for deep learning, modelling and understanding of spatio-/spectro temporal data, referred here as STDM [5]. A STDM has modules for: preliminary data analysis, data encoding into spike sequences, unsupervised learning of spatio-temporal patterns, classification, regression, prediction, model visualisation and knowledge discovery. The methodology is illustrated on benchmark data with different spatial/temporal characteristics, such as: EEG data for brain computer interfaces; fMRI data classification; personalised and climate date for individual stroke occurrence prediction [6]. The talk discusses implementation on highly parallel neuromorphic hardware platforms such as the Manchester SpiNNaker [7] and the ETH Zurich chip [8,9]. The STDM are not only significantly more accurate and faster than traditional machine learning methods and systems, but they lead to a significantly better understanding of the data and the processes that generated it. A STDM can be used to predict early and accurately events and outcomes through the ability of SNN to be trained to spike early, when only a part of a new pattern is presented as input data. New directions for the development of SNN and STDM are pointed towards a further integration of principles from the science areas of computational intelligence, bioinformatics and neuroinformatics [10,11].

References:
1. EU Marie Curie EvoSpike Project (Kasabov, Indiveri): http://ncs.ethz.ch/projects/EvoSpike/
2. Schliebs, S., Kasabov, N. (2013). Evolving spiking neural network-a survey. Evolving Systems, 4(2), 87-98.
3. Kasabov, N. (2014) NeuCube: A Spiking Neural Network Architecture for Mapping, Learning and Understanding of Spatio-Temporal Brain Data, Neural Networks, 52, 62-76.
4. Kasabov, N., Dhoble, K., Nuntalid, N., Indiveri, G. (2013). Dynamic evolving spiking neural networks for on-line spatio- and spectro-temporal pattern recognition. Neural Networks, 41, 188-201.
5. Kasabov, N. et al (2015) A SNN methodology for the design of evolving spatio-temporal data machines, Neural Networks, in print.
6. Kasabov, N., et al. (2014). Evolving Spiking Neural Networks for Personalised Modelling of Spatio-Temporal Data and Early Prediction of Events: A Case Study on Stroke. Neurocomputing, 2014.
7. Furber, S. et al (2012) Overview of the SpiNNaker system architecture, IEEE Trans. Computers, 99.
8. Indiveri, G., Horiuchi, T.K. (2011) Frontiers in neuromorphic engineering, Frontiers in Neuroscience, 5, 2011.
9. Scott, N., N. Kasabov, G. Indiveri (2013) NeuCube Neuromorphic Framework for Spatio-Temporal Brain Data and Its Python Implementation, Proc. ICONIP 2013, Springer LNCS, 8228, pp.78-84.
10. Kasabov, N. (ed) (2014) The Springer Handbook of Bio- and Neuroinformatics, Springer.
11. Kasabov, N (2016) Spiking Neural Networks: The machine Learning Approach, Springer, 2016

 
   

"Rapid testing, prototyping and validating new ideas thanks to: Data Lake, Azure Machine Learning and Azure Notebook"
Tomasz Kopacz
Microsoft

Abstract
During 45 minutes talk we will see how to use cloud based technologies to speed up developing a new ideas and concepts. In short – how to do RAPID prototyping in 2017. For sure, we will focus on:
- Really big data processing on Data Lake.
- Testing machine learning ideas on many large (and small) data sets.
- Develop and discuss Python / R applications using Azure Notebooks.
Be warned: this session will be a little bit technical!

 
   

"Decomposable Graphical Models in Industrial Applications: On Learning and Revision"
Rudolf Kruse
Faculty of Computer Science, University of Magdeburg, Germany

Abstract
Decomposable Graphical Models are of high relevance for complex industrial applications. The Markov network approach is one of their most prominent representatives and an important tool to structure uncertain knowledge about high dimensional domains. But also relational and possibilistic decompositions turn out to be useful to make reasoning in such domains feasible. In this talk we study how to generate the structure of the model from data as well as from background knowledge. A second important task in this context is the efficient revision of the model, when new data and knowledge become available. Here, the problem of handling inconsistencies is of utmost relevance for real world applications. We address these topics by presenting a successful complex application in automotive industry.

 
   

"System Modeling and Data Analytics - A Perspective of Information Granules "
Witold Pedrycz
Department of Electrical & Computer Engineering
University of Alberta, Edmonton Canada
and
Systems Research Institute, Polish Academy of Sciences
Warsaw, Poland


Abstract
The apparent challenges in system modeling and data analytics inherently associate with large volumes of data, data variability, and an evident quest for transparency and interpretability of established constructs and obtained results. We advocate that information granules play a pivotal role in addressing these key challenges. We demonstrate that a framework of Granular Computing along with a diversity of its formal settings offers a critically needed conceptual and algorithmic environment.

A suitable perspective built with the aid of information granules is advantageous in realizing a suitable level of abstraction and becomes instrumental when forming sound, practical problem-oriented tradeoffs among precision of results, their easiness of interpretation, value, and stability (as lucidly articulated through the principle of incompatibility coined by Zadeh). All those aspects emphasize importance of actionability and interestingness of the produced findings.

Special attention is paid to the construction of information granules and the talk tackles their design issue by emphasizing that the emergence of semantically sound granules has be justified by available experimental evidence. The rationale behind the emergence of information granules of higher type is offered and their unique role in realizing a hierarchy of processing and coping with a distributed nature of available data is presented. In system modeling, information granules are instrumental in the realization of granular models involving information granules of increasingly higher type. With this regard, we introduce concepts of granular spaces, viz. spaces of granular parameters of the models and granular input spaces, which play a pivotal role in granular models.

The detailed investigations are also reported for several selected classes of problems: (i) building granular auto-encoders in architectures of deep learning, (ii) realization of imputation mechanisms augmented by quantification of quality of imputed data, (iii) construction and analysis of hotspots, and (iv) carrying out knowledge transfer.

 
   

"COMPUTATIONAL INTELLIGENCE IN BIOMEDICAL ENGINEERING"
Ryszard Tadeusiewicz
AGH University of Science and Technology, Krakow, Poland

Ryszard Tadeusiewicz obtained his Master of Science degree with honors from the AGH University of Science and Technology in 1971 and started research in the areas of bio-cybernetics, control engineering, and computer science. In 1975 he was awarded by the Ph.D. degree, and in 1981 the degree of Doctor of Sciences. In 1986 he was appointed as associate professor and in 1991 full professor at the AGH University of Science and Technology. He has written over 800 scientific papers, published in prestigious Polish and foreign scientific journals, as well as in numerous conference proceedings - both national and international. Professor Tadeusiewicz has also written over 80 scientific monographs and books, among them several highly popular textbooks, which were adopted by dozens of Polish universities and had many editions. Prof. Tadeusiewicz supervised the total of 69 PhD students as the primary advisor at the AGH University of Science and Technology, Academy of Economics, and Collegium Medicum. In March 2002, Professor Tadeusiewicz was elected as Corresponding Member of the Polish Academy of Sciences (PAN) and in May 2012 he was elected as Full Member of PAN. He was three times elected as President of Cracow branch of PAN. In 1996, he was elected as Deputy Rector for Science of the AGH University of Science and Technology, and in January 1998 Rector of that University. He was re-elected again as Rector in 1999 and once more in May 2002 for the period 2002-2005. This makes him the longest-serving Rector of the AGH University of Science and Technology in Cracow. Details and most up-to-date information are available at the website http://www.Tadeusiewicz.pl.

Abstract
Biomedical engineering is new and fast developed area of scientific research and also very important element of up-to-date technology. Development of science and technology, which can serve doctors as a weapon against illnesses and death is especially worthy of support area of intellectual and technological activity. Seeking to bridge the gap between engineering and medicine, what is the main goal of biomedical engineering, we can use computational intelligence as very important tool for advance health care treatment, including diagnosis, monitoring, and therapy. Taking into account current state of art as well own research, author will present some illustrative examples of computational intelligence applications for more effective medical data analysis (for enhanced diagnosis). Moreover, interesting and inspiring examples of computational intelligence applications for therapy enhancement will be presented along with discussion of specific problems connected with communications between medical doctors and artificial (computer based) advisors. A very interesting part of computational intelligence used in biomedical engineering is development of intelligent medical devices. Some examples, including intelligent pacemakers, infusion pumps, the heart-lung machine, dialysis machines and cochlear implants, will be presented and assessed during the lecture. A very important part of computational intelligence application in biomedical engineering is telemedicine. Using modern sensors (including wearable ones) and applying current methods of wireless signal transmission we can collect a lot of medical data taken from many users. Practical use of these data (e.g. for selecting rare endangered persons, who need quick help, from big population of telemedically monitored people) need computational intelligence support. All mentioned above examples lead to conclusions, that both modern biomedical engineering need computational intelligence and developed computational intelligence tools can be implemented in biomedical engineering.