 |
Prof. Dr. Boris Delibašić
Boris Delibašić has been a full professor at the University of Belgrade – Faculty of Organizational Sciences since 2016. He was born in 1978. His research interests are related to data science, machine learning, business intelligence, multi-criteria decision analysis, and decision support systems. He is the vice EIC of the COMSIS journal. He is the deputy coordinator of the EURO (Association of European Operational Research Societies) working group on Decision Support Systems. He earned his PhD in 2007 at the University of Belgrade. He was a visiting lecturer at Friedrich Schiller University in Jena, Germany, from 2006 to 2011. He received the prestigious Fulbright scholarship for the 2011/2012 academic year at the Center for Data Analytics and Biomedicine, Temple University, Philadelphia, Pennsylvania, USA. He is often engaged as a consultant, helping companies utilize their data and decision-support system solutions. He has been the lead researcher on projects funded by various research agencies (Swiss National Science Foundation, German Academic Exchange Service, Office of Naval Research (USA), Serbian Ministry of Science). He has published more than 50 articles in impact factor journals and has an h-index of 21 (Google Scholar) and 15 (SCOPUS), with a total citation count of 1,860 (Google Scholar) and 873 (Scopus). He is fluent in English (C2), German (C2), and French (C1), and has solid knowledge of Italian (B1) and Russian language (B1). |
Talk title: From Data Science to Decision Support and Back!
Abstract: Data Science (DS) and Decision Support Systems (DSS) have long existed as both complementary and competing technologies. While DS has recently assumed a leading role by promising the automatic construction of DSS models, the foundational aim of DSS has been to make the decision-making process more informed and consistent without seeking to replace the human decision-maker. DSS models are traditionally designed to be correct, complete, consistent, comprehensible, and convenient—attributes that DS models do not always possess and whose potentials are often not fully understood within the DS community. In this talk, I will demonstrate how DS models can be integrated into DSS frameworks and how leveraging the strengths of DSS can enhance the effectiveness of DS models in achieving their objectives.
 |
Academician Janusz Kacprzyk
Janusz Kacprzyk is Professor of Computer Science at the Systems Research Institute, Polish Academy of Sciences, WIT – Warsaw School of Information Technology, AGH University of Science and Technology in Cracow, and Professor of Automatic Control at PIAP – Industrial Institute of Automation and Measurements in Warsaw, Poland. He is an Honorary Foreign Professor at the Department of Mathematics, Yli Normal University, Xinjiang, China. He is a Full Member of the Polish Academy of Sciences, Member of Academia Europaea, European Academy of Sciences and Arts, European Academy of Sciences, International Academy of Systems and Cybernetics (IASCYS), Foreign Member of the: Bulgarian Academy of Sciences, Spanish Royal Academy of Economic and Financial Sciences (RACEF), Finnish Society of Sciences and Letters, Flemish Royal Academy of Belgium of Sciences and the Arts (KVAB), Russian Academy of Sciences. National Academy of Sciences of Ukraine and Lithuanian Academy of Sciences. He was awarded with 8 honorary doctorates. He is a Fellow of IEEE (Life), IET, IFSA, EurAI, IFIP, AAIA, I2CICC, and SMIA. His main research interests include the use of modern computation computational and artificial intelligence tools, notably fuzzy logic, in systems science, decision-making, optimization, control, data analysis and data mining, with applications in mobile robotics, systems modelling, ICT etc. He authored 7 books, (co)edited more than 150 volumes, (co)authored more than 650 papers, including ca. 150 in journals indexed by the WoS. He was listed in 2020 and 2021 as”World’s 2% Top Scientists” by Stanford University, Elsevier (Scopus) and ScieTech Strategies and published in PLOS Biology Journal. He is the editor-in-chief of 8 book series at Springer and of 2 journals and is on the editorial boards of ca. 40 journals. He is President of the Polish Operational and Systems Research Society, Past President of the International Fuzzy Systems Association, and is a member of the Adcom (Administrative Committee) of the Computational Intelligence Society of the IEEE, and a member of the Board of Governors of the Systems, Man and Cybernetics Society of the IEEE. |
Talk title: Towards AI-assisted/Powered Smart Environments: from Techno-Centric to Human-Centric and Value-Centric Approaches
Abstract: The concept of the so-called smart environments is recently attracting much attention both in science, R&D and even media. The smart environments are basically "small worlds" that constitute a collection of sensors, computers and humans – both individuals and social groups of various sizes - who are synergistically integrated to better complete some tasks and fulfil goals, implement policies, meet expectations by various stakeholders, etc. We are concerned with environments in which the human stakeholders are crucial, and they involve individuals, social groups, enterprises, organisations or even the society as a whole. The problem is to develop and implement some automated agents who can help the human stakeholders develop, plan and finally implement strategies, policies and operations which are crucial for the particular human stakeholders for dealing with problems they face. We assume that the availability and access of various broadly perceived sensors, controllers, etc. is constant and pervasive. Recently, the use of tools and techniques that are AI (artificial intelligence)- assisted, AI-powered or even AI-enabled is strongly advocated in virtually all fields, also here. Our main interest is the general human-centric direction in smart environments, which is a new way of extending the traditional techno-centric perspective, i.e. just concerning sensors and other inanimate tools and techniques, of the smart environments by assuming that the human being is the key player (stakeholder, actor) as in virtually all complex real-world problems. Such a perspective implies a very wide and complex agenda of research and implementations combining an active and proactive involvement in the operation of the smart environment, and incentives for the participation of the humans. One of the aspects to be accounted for is a need to find proper relations between the egocentric and (social) value-oriented views. Some inspirations from the human-in-the-loop and society-in-the-loop paradigms can here be employed. We show how artificial intelligence (AI) can provide tools and techniques to develop new decision-making and reasoning models via the so-called AI-assisted perspective. Then, to deal with more complex decision and reasoning problems, we show the use of the so-called AI-powered/enabled approach, notably in the decision support systems perspective. We emphasise difficulties and challenges faced by the human stakeholders in their role as cognitive partners of automated agents in advanced smart environments, e.g. due to human cognitive biases. Some examples of the use of our approach for critical infrastructure planning are shown.
 |
Prof. Dr. Bożena Kostek
Bożena Kostek is a professor at the Gdansk University of Technology, Poland. She is a corresponding member of the Polish Academy of Sciences, a Fellow of the Acoustical Society of America, and of the Audio Engineering Society. She published more than 600 scientific papers in journals. Under her guidance, 25 Ph.D. students supported their doctoral theses and 320 M.Sc. and Eng. works. She is the recipient of many prestigious awards, including two 1st prizes from the Prime Minister of Poland, several prizes from the Minister of Science, and an award from the Polish Academy of Sciences. She was Editor-in-Chief of the Journal of Audio Engineering Society (2011-2020) and Archives of Acoustics from 2008 to 2012. She was also Guest Editor of the J. Intelligent Information Systems, JAES (J. Audio Eng. Soc.) and J. Acoustical Society of America. She serves as a reviewer of many scientific journals (e.g., IEEE Transactions on Audio, Speech and Language Processing). She led many research projects sponsored by the Polish National Centre for R&D and the National Science Centre and received more than 20 patents. She also authored or coauthored some technological solutions (40). Her research activities are interdisciplinary. However, the main research interests focus on cognitive bases of hearing and vision, music information retrieval, musical acoustics, studio technology, Quality-of-Experience, human-computer interaction (HCI), as well as applications of soft computing and computational intelligence to the mentioned domains. Prof. Bożena Kostek was elected three times as Vice President of the Audio Engineering Society for Central Europe ( 2003-2005, 2005-2007, and 2009-2011), twice Governor of the Board of Governors of the AES ( 2007-2009 and 2011-2013), and she was a member of the AES Executive Committee for nine years. |
Talk title: Perception-Driven Augmented Reality and Its Role in Machine Learning
Abstract: This study explores the notion of how perception-driven augmented reality (AR), as processed by Artificial Intelligence (AI) systems, can drive advancements in technologies that play a role in the field of machine learning. However, the title of this presentation is only clear when the traditional approach to AR, which refers to an interactive experience that combines the real world and computer-generated 3D content, is reframed from the lens of "looking closer and in more detail into a subject to be processed by AI.” Hence, this is a way for AI to "zoom in" on and process the world around us to a granular degree. There are many examples of such approaches, one of them being perceptual computing, allowing human-computer interfaces to understand and respond to inputs such as voice, textual information, image, and gestures enhanced by emotions. The granularity mentioned earlier refers to the expertise derived from human perception. Despite the fact that deep models, including transformers, can achieve high-level quality performance without human intervention, it does not change the fact that the resources on which these models are trained use all this information inscribed in the data. It should also be remembered that what is natural to humans when analyzing, e.g., medical data, such as zooming in on a specific feature or region of interest and then describing it in a natural language, may not yet be possible for an AI model to achieve that, specifically when it concerns underrepresented languages in the medical domain. In this study, it will be shown that an augmented approach to intelligent speech processing may resolve some of the problems involved when medical data is involved. One may envision structured forms and interfaces provided to a physician or medical assistant when a patient’s data should be filled in to evolve into improved structures based on contextual awareness and feedback from the process. However, the basic problem that remains so far unresolved is more common, i.e., when the environment interacts with the healthcare provider during the process of recording the details of a patient visit. In such a case, sensing an environment and deriving its acoustic and noise properties should be analyzed with a granular level of detail based on human expertise and then processed by an AI model to gain human-like perception. This study was supported by the Polish National Center for Research and Development (NCBR) project: “ADMEDVOICE-Adaptive intelligent speech processing system of medical personnel with the structuring of test results and support of therapeutic process,” no. INFOSTRATEG4/0003/2022.
 |
Prof. Dr. Elijah Liflyand
Experience in applied mathematics, mathematical statistics, processing of information, and computer programming. Committees for PhD evaluation in various countries; Committee for nostrification in Israel of foreign PhD degrees in natural and technical sciences. Member of the Israel Mathematical Union, American Mathematical Society, and European Mathematical Union. List of publications is more than 180 items - 3 books among them. Associate Editor: Journal of Fourier Analysis and Applications; Journal of Mathematical Sciences, Series A; Integral Transforms and Special Functions. |
Talk title: Certain Computational Aspects in Approximation and Harmonic Analysis
Abstract: In applications of approximation and harmonic analysis, where numerical methods come into play, sharpness and optimality of relations submitted for calculations are often of crucial importance. Even better if such relations are asymptotic, when convergence of Fourier expansions is studied (just approximation, signal processing, etc.), the so-called Lebesgue constants are an object whose behaviour must be thoroughly analyzed. The situation becomes even more complicated when linear means of Fourier expansions are involved rather than partial sums. For that case, a result due to my late friend Eduard Belinsky has been for decades the best known. However, it was in the form of bilateral estimates, where many additional terms had to be estimated. We have recently succeeded to shape it in an asymptotic way. In particular, if the asymptotics is "genuine" (which may be achieved by the appropriate choice of a function generating summability method), that is, the remainder term is of really better behaviour, one has to concentrate theoretical and numerical efforts only on the leading term. Concrete examples illustrate the latter.
 |
Prof. Sanda Martinčić-Ipšić
Sanda Martinčić-Ipšić obtained her B.Sc. degree in Computer Science from the University of Ljubljana Faculty of Computer Science and Informatics, and her M.Sc. degree in informatics from the University of Ljubljana, Faculty of Economy. In 2007 she obtained a Ph.D degree in Computer Science from the University of Zagreb, Faculty of Electrical Engineering and Computing. Dr. Martinčić-Ipšić currently works as full professor of Computer Science at the University of Rijeka, Faculty of Informatics and Digital Technologies. She is a leader of Language Networks research group, and the head of Laboratory for Natural Speech and Language processing at Center for Artificial Intelligence and Cybersecurity at University of Rijeka. Her research interests include natural language processing, complex networks, data science and data analytics. She has published more than 100 scientific papers and books. |
Talk title: From Text to Insight: Enhancing Relation Extraction in Climate Change Research
Abstract: Global warming and climate change have profound and far-reaching effects on global ecosystems, weather patterns, sea levels, and human societies, constituting a critical threat to the planet's biodiversity and the prospect of a sustainable future. Simultaneously, the volume of climate change data is rapidly increasing, particularly in published scientific publications. The automated processing of information from unstructured textual data is crucial, with a primary focus on the natural language processing task of information extraction and, more specifically, its subtask, relation extraction. Relation extraction involves identifying relationships between entities within sentences, paragraphs, or larger text units, aiming to automatically generate machine-interpretable data collections that capture entities, their relationships, and associated attributes. This talk addresses the challenge of extracting named entities and relations from scientific publications from renowned journals in the climate change domain. Firstly, the statistics of the collected dataset will be presented along the problems encountered in data preprocessing. Secondly, the domain adaptive pretraining of the SciBERT and Climate(Ro)BERT(a) models and from scratch training of CliReBERT and CliReRoBERTa models will be elaborated. The discussion will focus on the model architectures and training parameters used, highlighting the advantages and disadvantages of domain adaptive pretraining compared to training from scratch. Thirdly, the task of extracting relations and named entities in the climate change domain will be elaborated upon, presenting results on LLM-enabled relation annotation and discovery. These results will be used to train all pretrained models (i. e. BERT and RoBERTa) to supervised relation extraction downstream task. Finally, the plan for constructing a knowledge graph from extracted relations in the climate change domain will be discussed.
 |
Andrius Januta
Andrius Januta is a cybersecurity technical manager at Nord Security. His responsibilities encompass designing, implementing, and maintaining the company’s cybersecurity strategy, including deploying and managing advanced security tools to protect sensitive data. Over the past five years, he has been an active participant in Lithuania’s cybersecurity and defense exercise, Amber Mist, where he served as a core member of both the Red Team and the technical cyber range development team. Additionally, he has been a regular participant in various cybersecurity exercises organized by the NATO Cooperative Cyber Defence Centre of Excellence (CCDCOE). |
Talk title: Adversarial Attacks on AI: Understanding and Securing Machine Learning in Cybersecurity
Abstract: Artificial Intelligence (AI) and Machine Learning (ML) have become essential tools in modern cybersecurity, but these models themselves are vulnerable to a wide range of attacks. One of the most serious threats is adversarial attacks, where malicious actors manipulate the inputs of ML models to produce incorrect or harmful outputs. The presentation will explore the primary vulnerabilities of AI and ML models that pose a threat to cybersecurity. The main focus will be on understanding how adversarial attacks work, the consequences they can have for cybersecurity applications, and how these attacks can be used to facilitate malicious activities such as manipulating machine learning outcomes. We will also examine the threats of prompt hacking, where AI models become susceptible to deceptive queries.
 |
Assoc. Prof. Dr. Dmitri E. Kvasov
Associate Professor in Numerical Analysis, DIMES, University of Calabria, Rende (CS), Italy. Italian National Scientific Habilitation as Full Professor in Numerical Analysis (2018–2027) and in Operations Research (2021–2030). Education: Ph.D. in Operations Research (05/2006), Department of Statistics, University of Rome "La Sapienza", Italy. Candidate (Ph.D.) of Physico-Mathematical Sciences (12/2016), "Lobachevsky" University of Nizhny Novgorod, Russia. Graduated, with honours, in Information Systems (06/2001), Faculty of Computational Mathematics and Cybernetics, "Lobachevsky" University of Nizhny Novgorod, Russia. Graduated, with honours, in Computer Systems Engineering (04/2001), Engineering Faculty, University of Calabria, Italy. Research interests: Numerical analysis; Continuous global optimization and applications; High-performance and Infinity computing. List of papers includes more than 130 items (among them: 2 research books). Research Interests: Continuous global optimization and applications; high-performance and infinity computing |
Talk title: Advanced Global Optimization Techniques and Their Applications
Abstract: In many simulation-based applications of optimization, the objective function can be multi-extremal and non-differentiable, thus precluding the use of descending schemes with derivatives. Moreover, the function is often given as a black-box and, therefore, each function evaluation is an expensive operation with respect to the available computational resources. Derivative-free methods can be particularly suitable to address these challenging problems studied in the framework of global optimization and can be deterministic or stochastic in nature. A numerical comparison of these two groups of methods is interesting for several reasons and has a notable practical importance. In the presentation, the methods of these two groups are considered and their applications (including the field of machine learning) are briefly examined.
 |
Rajesh Sharma is presently (from January 2021 - present) working as an Associate Professor and is leading the Computational Social Science Group (https://css.cs.ut.ee/) at the Institute of Computer Science, University of Tartu (UT), Estonia. Rajesh also holds an adjunct professor position at IIT Delhi and IIT Ropar, both in India and Lakehead University, Canada. His group works on problems related to understanding societal issues such as misinformation, hate speech, segregation, mental health, and users' behaviour using digital traces such as financial transactions, mobile call data records and, more importantly, social media data. Group often applies techniques from AI, NLP and, most importantly, network science/social network analysis. Rajesh has co-authored over 80 research articles and regularly publishes in journals such as 1)IEEE Transactions on Network Science and Engineering, 2) IEEE Transactions on Computational Social Systems, 2) Journal of Pervasive and Mobile Computing, 3) International Journal of Data Science and Analytics, 4) Plos One, 5) Journal of Social Network Analysis and Mining, and international conferences, such as AAAI ICWSM, PerCom, CIKM, ACM WebSci, COLING, NAACL, IJCNN, ASONAM, IEEE Big Data. Earlier, Rajesh joined the University of Tartu in August 2017 and worked as a senior researcher till December 2020. From Jan 2014 to July 2017, he has held Research Fellow and postdoc positions at the University of Bristol, Queen's University Belfast, UK and the University of Bologna, Italy. Before that, he completed his PhD at Nanyang Technological University, Singapore, in December 2013. He has also worked in the IT industry for about 2.5 years after completing his master's from the Indian Institute of Technology (IIT), Roorkee, India. |
Talk title: Social Media Analytics for Making Online Platforms Safer!
Abstract: Societal issues such as misinformation and abusive speech have become a menace in recent times. Individuals have started (mis)using online social media platforms for spreading misinformation and hate speech, primaritly due to the anonymity provided by these platforms and in the name of free speech. These are critical issues as they can create riots in a society, leading to loss of property and lives. In this talk, Rajesh will touch upon various societal issues (mentioned earlier), which can be studied using several data science techniques. The digital traces, if studied responsibly and scientifically, can help to mitigate this menace in society and can make these platforms safer for the society.
 |
Professor Yeh obtained his PhD degree in Electrical Engineering from National Taiwan University in 2004. Subsequently, he became a faculty member in the Department of Biomedical Engineering and Environmental Sciences at National Tsing Hua University (NTHU), Taiwan in 2005, where he excelled in academics, teaching, and social services. Professor Yeh's research focuses on the application of ultrasound technology in the biomedical field, with a particular emphasis on three main areas: (1) tornado-inspired acoustic vortex technology for advanced medical applications, (2) ultrasonic neuromodulation and sonogenetics of brain disorders, and (3) precision medicine applications of ultrasound. He has received numerous awards, including the Wu Dayou Memorial Award from the National Science and Technology Council of Taiwan (NTSC), the Outstanding Research Award twice by NTSC, and the Future Science and Technology Award twice by NTSC. He currently holds the position of chair professor/ chief secretary at NTHU and is a fellow of International Academy of Medical and Biological Engineering (IFMBE fellow) and a senior member of the IEEE Society as well as an editor for three SCI journals. Professor Yeh has published over 160 papers and has presented over 200 proceedings abstracts/papers at seminars and conferences. Professor Yeh has extensive experience in implementing industry-university cooperation and has obtained 20 patents. He has also started two companies that specialize in ultrasound contrast agents (Trust Bio-sonics, 2013) and catheter-based ultrasound (SoundJet Medical, 2022). In terms of social services, while serving as the convener of the biomedical engineering division at NTSC of Taiwan, he promoted the development of innovative medical devices and made contributions to bilateral international research collaboration and academic conference exchanges. |
Talk title: Trapping and manipulating drug-loaded microbubbles by acoustic vortex tweezers
Abstract: Microbubbles (MBs) can be pushed through blood circulation under radiation force guidance and facilitate the drug adhesion via cavitation. However, the retention and accumulation of MBs on the target site is typically unstable under flow conditions, particularly in the regions of endarterial region and thrombosis. In this study, we propose the acoustic vortex tweezers (AVT) precisely collecting MBs at specific locations under different flow conditions and ultrasound parameters. Owing to the features of long working distance (>10 mm) and single beam configuration, the AVT become feasible in vivo applications. Moreover, the AVT trapping drug-loaded MBs perform drugs accumulation at specific site within blood vessel and B-mode images can see the manipulating process of MBs. The AVT trapping process was optically observed in a 200-μm capillaries mimic tube and acoustically monitored by a 7 MHz B-mode imaging in a tissue mimic phantom. Self-made MBs (sizes of 1.2 μm) were injected with a 1.7 cm/s flow velocity. The AVT was realized by a 4-element customized transducer with 90-degree phase different in each adjacent element (frequency: 3.1 MHz, pressure: 800 kPa, duty cycle: 9.6%). When the AVT was applied, primary radiation force displaced MBs perpendicularly from the center streamline and the secondary radiation force make the MBs aggregation to reach size of 12-14 μm simultaneously. The AVT collected the attached clusters toward the potential-well to form a big cluster 31.1 μm and continued to retain in the flow without lost. The trapping size of 240μm was measured by the maximum pressure gradient. The retaining big cluster can be detected by B-mode imaging resulting to 1.06 mm speckle pattern. These results suggested that AVT are useful in precise retention of MBs under intravascular conditions and the surveillance ability can ensure process safety. Furthermore, MBs signals within mouse capillaries could be locally improved 1.7-fold and the location of trapped MBs could still be manipulated during the initiation of AVT. The proposed AVT technique is a compact, easy-to-use, and biocompatible method that enables systemic drug administration with extremely low doses.