Invited Speakers

Keynote Speakers

Handling Imperfect 'Big' Data in Decision Support Systems

With ever increasing data volumes, database and information management systems face new challenges. Four important characteristics of `Big' data, commonly known as the four V's, are huge Volume, large Variety in data formats, high Velocity and Veracity. Veracity refers to the trust one has in the data that are being used. In this talk we want to address novel techniques in computational intelligence for the proper handling of veracity problems. Topics being presented include the adequate presentation and handling of data imperfections, data quality assessment and result visualization in (geographic) decision support systems. As case study, aspects from the Belgian Federal project Transnational and Integrated Long-term marine Exploitation Strategies (TILES) will be presented.

Prof. Guy De Tré is currently Head of the Database, Document and Content Management (DDCM) research group, belonging to the Department of Telecommunications and Information Processing of the Faculty of Engineering and Architecture at Ghent University. He finished secondary school at the KA Ninove and received a B.Sc. in Mathematics and a M.Sc. in Computer Science from Ghent University where his master thesis was on optimizing primary disk organization for databases. He also completed a PhD in Applied Sciences at Ghent University on the handing of imperfect information in database systems. In 2002 he was visiting researcher at the Institut de Recherche en Informatique de Toulouse - IRIT (Université Paul Sabatier, Toulouse, France) and the Systems Research Institute (Polish Academy of Sciences, Warsaw, Poland).

In 2004 Guy became tenured docent (assistant professor) at Ghent University and co-established DDCM. In 2010 he became head docent (associate professor). Previously, he worked as consultant in data and knowledge engineering for private companies like NMBS and Fabricom.
Adaptive mobile robots for environment and agriculture

Robots are more and more present in everyday life to achieve difficult, repetitive and harmful tasks. They are particularly popular in industrial manufacture, such as automotive, and works in assembly line. In such areas, robots are particularly performant and reliable, since the environment is well known and structured. Nevertheless, outdoor robotics is still emergent, even in the field of autonomous cars, since unforeseen situations are ofen encountered, and the structure, such as road signs, may be missing or difficult to sense. This is particularly true in off-road contexts, where a structure is not always available and the conditions of motion are varying deeply and cannot be predicted precisely. In order to be efficient and help people reduce the work hardness in applications such as agriculture, outdoor mobile robots have to adapt their behavior with respect to these conditions and the task to be achieved.

This talk proposes an overview of advanced control technique allowing to take into account the varying and dynamic phenomena disturbing the robot dynamics. In particular, adaptive and robust control techniques allowing to estimate on-line such variations will be presented. They permit to improve the accuracy of mobile robot navigation despite disturbances. Nevertheless, motion control does not address the risk controllability loss (spin around, roll-over) which can quickly occurs in uneven and slippery grounds. Beyond obstacle avoidance, which is deeply studied in mobile robotics, the adaptation of robots behaviour in off-road context must also account for the robot integrity, and possibly depreciated perception (because of light conditions, dust of rain).

Dr. Roland Lenain is a Research Director at IRSTEA (Institut national de recherche en sciences et technologies pour l’environnement et l’agriculture - National Research Institute for Science and Technology for the Environment and Agriculture) at Antony - France. Dr. Lenain is in charge of the Romea Research group, which develops robots that have the capacities to work in natural environments. In particular, he oversees research into the modeling and control of these robots in order to adapt their behavior to the situations encountered.
Why Swarm Intelligence seems so appealing for complex problem solving

Nature has been inspiring “computational tool makers” for a long time. This, especially when swarms and colonies can serve well as metaphors for tackling complex problems, as their ability to cope with large dimensional search spaces of highly non-monotonic nature, their flexibility and scalability, and rather low computational cost presents Swarm Intelligence as an appealing set of tools.
In 2008, Bastos Filho and Lima Neto proposed a new metaheuristic in the fast-growing family of swarm intelligence techniques, namely, Fish School Search (FSS). In FSS, the school collectively “swims” (searches) for “food” (candidate solutions) in the “aquarium” (search space). Similarly, to PSO (Particle Swarm Optimization) or GA (Genetic Algorithms), the search guidance in FSS is driven by the merit of individual members of the population and the weight of each fish acts as a factual-memory of its individual success. In contrast with PSO, the weight can obviate the need to keep a log of best positions visited as well as any other topological information. As opposed to GA, the actual location of each fish directly substitutes the need of a chromosome. As for the social reasoning, the barycenter of the whole school can guide expansion and contraction of the school, automatically evoking exploration and exploitation when necessary. In other words, the quality of the search can be inferred from regions where larger ensembles of fish are located (and vice-versa).
In this, some reflections on research avenues of Swarm Intelligence will be offered, focusing on aspects that are sometimes disregarded and might be offered why Swarm Intelligence seems so appealing for complex problem-solving. Along the presentation, FSS and some of its variations will be used to illustrate the key-points that are going to be commented upon.

Prof. Fernando Buarque de Lima Neto has a Ph.D. in Artificial Intelligence from University of London (2002), Diploma of Imperial College London on Artificial Neural Networks (2002), Masters in Computer Science from Federal University of Pernambuco (1998) and a degree in Computer Science Catholic University of Pernambuco (1990). He also studied Philosophy of Science, Neuroscience, and Administration. In 2012 he was at INRIA (Paris-Rocquencourt) as ‘Professeur invité’ and in 2015 he was on sabbatical leave at Westfälische Wilhelms-Universität Münster, Germany. Buarque has produced over 200 bibliographic products (e.g. articles, projects and books). He is Associate Professor at the University of Pernambuco, IEEE Senior Member, Alexander von Humboldt Fellow and Accredited Researcher by the Brazilian Council of Research. Currently, he has four international appointments with: (i) Texas A&M (Adjunct Professor), (ii) Florida Tech (Graduate Faculty), (iii) University of Johannesburg (Visiting Professor), and (iv)University of Münster (Visiting Professor). His current research focuses at: (1) Computational Intelligence (Evolutionary, Social and Hybrid Metaheuristics), (2) Complex and Stochastic Modeling/Simulation, and (3) Intelligent and Semiotic Decision Support Systems.

Fernando Buarque is a member of several scientific organizations among them the Brazilian Computer Society (SBC), the Brazilian Society of Computational Intelligence (SBIC, past president), the Computational Intelligence Society (IEEE-CIS) and the Systems, Man & Cybernetics Society (IEEE- SMC). In recent years he has committed himself to internationalization at large, when he co-chaired/steered several international conferences such as BRICS-CCI 2013(Recife-Brazil), BRICS-CCI 2015(Beijin-China), LA-CCI 2014(Bariloche-Argentina), LA-CCI 2015(Curitiba-Brazil), IEEE LA-CCI2016(Cartagena-Colombia), IEEE LA-CCI2017(Arequipa-Peru) to name a few.
A Submodular Approach to Power System Stability

In this talk, I will discuss on-going research in developing scalable control
algorithms that provide verifiable guarantees on power system stability. Power system stability must be enhanced in order to realize large-scale clean energy systems, due to the stability challenges posed by unpredictable renewable energy sources such as wind and solar. Since the power system is inherently complex and nonlinear, current state of the art methods, developed by groups such as the North American Synchrophasor Initiative (NASPI), rely on heuristics informed by domain expertise and “rules of thumb”, which are then verified by simulation studies. Our
insight is that many of the relevant stability problems exhibit inherent submodular structure arising from the physical invariants of the power system. Submodularity is a diminishing returns property that has extensive applications in machine learning but was not explored for power systems prior to our work. The advantage of the submodular approach is that it leads to computationally efficient algorithms with provable optimality guarantees.

We will discuss two problems in this talk, the first being the development of
submodular algorithms for voltage stability, in particular the problem of selecting which buses should inject reactive power in order to ensure that a desired voltage level is achieved at each bus of the system. We will show that widely-used metrics for quantifying the level of voltage stability are in fact submodular functions of the set of buses injecting reactive power. In the second problem, we investigate small-signal stability, which consists of mitigating unstable oscillations between different geographic regions of the power system. Recently, wide-area control schemes, in which a centralized controller receives measurements from different regions, computes control actions, and transmits supplementary damping control signals to distributed generators have been proposed to enhance small-signal stability. These schemes, however, rely on choosing a set of Phasor Measurement Units (PMUs) to send measurements, as well as choosing a set of generators to participate in the control. We formulated the problem of selecting minimum-size sets of generators and measurements in order to guarantee that all unstable modes are stabilized.

Prof. Linda Bushnell is a research associate professor in the Electrical Engineering Department of the University of Washington. She received a Ph.D. in electrical engineering and an MA in math from UC Berkeley, and both an MS and a BS in electrical engineering from the University of Connecticut. She also holds an MBA from the UW Foster School of Business. Her research interests include networked control systems and secure-control.

Bushnell is a recipient of the US Army Superior Civilian Service Award, NSF ADVANCE Fellowship and IEEE CSS Recognition Award. She was elected a Fellow of the IEEE for her contributions to networked control systems. She has been a member of the IEEE since 1985, a member of the IEEE CSS since 1990 and a member of the IEEE Women in Engineering since 2013. For IEEE CSS, she currently is a Member of the Board of Governors, a Distinguished Lecturer, a member of the Women in Control Standing Committee, a member of the TC Control Education, a member of the History Committee and the Liaison to IEEE Women in Engineering. For the American Automatic Control Council (AACC), she currently is the Treasurer and a Member of the Technical Committee on Control Education.
WASTCArD: Wrist and arm sensing technologies for cardiac arrhythmias detection in long term monitoring

Abnormal heart rhythms are a major cause of cardiovascular disease and death in Europe. Sudden cardiac death accounts for 50% of cardiac mortality in developed countries; ventricular tachycardia or ventricular fibrillation is the commonest underlying arrhythmia. In the ambulatory population, atrial fibrillation is the commonest one, and is associated with increased risk of stroke and heart failure, particularly in the aged population. If arrhythmias are detected at an early stage of heart disease, appropriate treatment can be effective, reducing disability and death. However, in the early stages of disease these may be transient, lasting only a few seconds, and thus difficult to detect.

The proposed joint research project through staff exchange activities, will investigate enabling technologies for non-invasive recording heart rhythm during long periods of time (>36h), using a wrist or arm wearable device with novel ECG sensing techniques and embedded real-time cardiac arrhythmia detection processes. The problem of extracting the far-field heart electrogram signal from noise components will be addressed using smart denoising algorithms. The project will impact by establishing a successful international (specifically with Colombia, Venezuela and Brazil) and intersectoral partnership for the development of new technologies addressing a significant cardiovascular healthcare problem. These technologies will be suitable for integration into current e-Health and cardiac information systems, and will impact on healthcare costs reduction by improved efficiency in the diagnosis and early treatment of cardiac disease.

Prof. Omar Escalona is a Professor in the School of Engineering and a member if the Engineering Research Institute. He is Director of the Centre for Advanced Cardiovascular Research (CACR) at Ulster University. His ongoing research and impact activities are on: atrial fibrillation solutions with improved QALY outcome difference; highly efficient cordless energy supply systems for artificial hearts and VADs; and Connected-Health enabled cardiovascular healthcare services. Since 1986, he has published more than 120 articles in prestigious International Journals and Conference Proceedings, all in the area of cardiovascular research. In 1994, he submitted his first successful
European academic networking grant proposal in the ALFA Programme (Project Alfa-Beta) and has had subsequent EU funding for his research since including a recent H2020 MRCA-RISE award entitled WASTCArD.

Professor Escalona has BSc Hons Degree in Electronic and Electrical Engineering from University of Surrey (1980); an MSc from University Simón Bolívar (USB), Venezuela (1986) and a PhD at Ulster University (1993). The latter was in the area of high-resolution ECG diagnostic techniques, which resulted in his first innovation patent: “Analysis of Heart Waveforms”, US Patent #5694942, 9/Dec/1997.
Big Data Analytics using Deep Learning and Information Theoretical Learning: Applications to Astronomy

Astronomy is facing a paradigm shift caused by the exponential growth of the sample size, data complexity and data generation rates of new sky surveys. To cope with a change of paradigm to data-driven science new computational intelligence, machine learning and statistical approaches are needed. In this talk I will present two main applications. The first is to discriminate periodic versus non-periodic light curves, and then estimate the period of the periodic ones. Light curves are one-dimensional time series of the brightness of a star versus time. We have developed several methods based on the correntropy function (generalized correlation using information theoretical learning concepts), which outperforms conventional approaches. Results using 32.8 million light curves will be presented. Interestingly, some of these techniques can be applied to other problems such as sleep EEG analysis, and I will present preliminary results on this topic too.

The second application is the automated real-time transient detection in astronomical images. The aim is to achieve real-time detection of supernovae and other transients with the Dark Energy Camera. A novel transient detection pipeline was developed. We have been applying convolutional neural nets (deep learning) to discriminate between true transients and bogus transients, among other techniques, e.g non-negative matrix factorization combined with random forests. Results using 1.5 million images will be presented. The new pipeline was successfully tested online in February 2015 finding more than 100 supernovae in a few days of telescope observation.

Prof. Pablo A. Estévez received his professional title in electrical engineering (EE) from Universidad de Chile, in 1981, and the M.Sc. and Dr.Eng. degrees from the University of Tokyo, Japan, in 1992 and 1995, respectively. He is a Full Professor with the Electrical Engineering Department, University of Chile, and former Chairman of the EE Department in the period 2006-2010.

Prof. Estévez is one of the founders of the Millennium Institute of Astrophysics (MAS), Chile, which was created in January 2014. He is currently leading the Astroinformatics/Astrostatistics group at MAS. He has been an Invited Researcher with the NTT Communication Science Laboratory, Kyoto, Japan; the Ecole Normale Supérieure, Lyon, France, and a Visiting Professor with the University of Tokyo.

Prof. Estévez is an IEEE Fellow. He is currently the President of the IEEE Computational Intelligence Society (CIS) for the term 2016-2017. He has served as IEEE CIS President-elect (2015), CIS Vice-president of Members Activities (2011-2014), CIS ADCOM Member-at-Large (2008-2010), CIS Distinguished Lecturer (2006-2011) and as an Associate Editor of the IEEE Transactions on Neural Networks (2007-2012).

https://www.researchgate.net/profile/Pablo_Estevez

Tutorial Sessions

A Pragmatic Review of the Cryptographic Concepts from the Theory to Practice

Francisco Bolaños is the Director of the master program Maestría en Auditoría de Tecnologías de la Información at UEES. He teaches cryptography and ethical hacking at the graduate level. His researcih areas are: cryptography, information security behavior and assessment tools.

In the educational field, he studies pedagogical ways to teach cryptography and ethical hacking to IT auditors based on a bussiness strategic approach. He has been a presenter in academic conferences held in Washington DC, Madrid, Berlin and Bologna. Futhermore, he is an alumni of the Electronic and Telecomunications Research Institute (ETRI) in Daejeon, South Korea.
Python - A Crash course for Communication Engineers

Marcel Jar is a professor at the School of Information and Communications Technology at Seneca College of Applied Arts and Technology, Toronto, Canada.

Marcel has a PhD in Telecommunications from the University of Alberta, and has worked for the past 12 years designing, simulating, and implementing communication systems in a myriad of different programming languages (C/C++, Matlab/Simulink, Python, VHDL) and technologies Microcontrollers, FPGA boards, Raspberry PIs, Beagle Bone, Arduino).
IBM Watson

Iván Sáenz Flor is an Analytics Architect at IBM Ecuador. He has worked with technology solutions for the private and public sectors during the last 14 years. He holds a B.S. in Computer Systems from Escuela Politécnica Nacional (2004). He was responsible in 2006 of the implementation of the business logic for the "Cuenta de Cesantía" for the Social Security Institute in Ecuador (IESS). He has also been the architect in chief of the implementation of the Enterprise Services Bus of the biggest bank in Ecuador. Since 2014 he works for IBM focused at looking for new ways to transform life in Ecuador through the use of technology.
Web Data Analytics

Carmen Vaca is full time professor at Escuela Superior Politécnica del Litoral. She is an expert in spatial data mining and social network analysis. She got her Ph.D. in Computer Science from Politecnico di Milano, and her Msc. in Computer Science from Syracuse University.

Carmen Vaca has performed research with the Max Planck Institute for Software Systems in Germany, and with Yahoo in Spain. She holds several certifications related to Machine Learning and Data Science.
Cardiovascular Engineering: challenges in electrical assessment, intervention and assistance of the heart.


Prof. Omar Escalona is a Professor in the School of Engineering and a member if the Engineering Research Institute. He is Director of the Centre for Advanced Cardiovascular Research (CACR) at Ulster University. His ongoing research and impact activities are on: atrial fibrillation solutions with improved QALY outcome difference; highly efficient cordless energy supply systems for artificial hearts and VADs; and Connected-Health enabled cardiovascular healthcare services.

Every heart beat involves electrochemical flow of charged particles which may be sensed on the body surface for assessing the heart’s state of health or any arrhythmic abnormality through multichannel ECG techniques, including challenging body surface cardiac mapping techniques in bipolar far-field body surface locations. Reversely, some heart abnormalities can be terminated by means of electrical intervention in various challenging and effective ways. Finally, overcoming the latter cardiac electrical intervention challenges may provide the long awaited technology solutions for sustainable wireless electrical power supply for the mechanical assistance of the weak and failing heart.
Data Acquisition Systems DAQ and Computer Based Instrumentation applied to electronic circuits and control systems


Santiago Orellana is the Instrumentation and Automation Manager of DataLights, a National Instruments Distributor in Ecuador. Mr. Orellana is a professional instructor certified by National Instruments. He has over twenty years of experience working with National Instruments solutions as a consultant, applications developer and trainer.

The tutorial will introduce the computer based instrumentation and the Data Acquisition Systems DAQ. Participants will have the opportunity to work with the simulation of electronic systems using Multisim, the virtualization of instrumentation, such as: oscilloscopes, function generators and Bode analyzers. Participants will also work in advanced circuit analysis and spectrum analysis.