Prof. Plamen P. Angelov

Lancaster University, UK

Title: Explainable AI through Interpretable Deep Rule-based Learning

Abstract: We are witnessing an explosion of data (streams) being generated and growing exponentially. Nowadays we carry in our pockets Gigabytes of data in the form of USB flash memory sticks, smartphones, smartwatches etc. Extracting useful and human-intelligible/understandable information and knowledge from these big data streams is of immense importance for the society, economy and science. The mainstream Deep Learning quickly became a synonymous of a powerful method to enable items and processes with elements of AI in the sense that it makes possible human like performance in recognising images and speech. However, the currently used methods for deep learning which are based on neural networks (recurrent, belief, etc.) is opaque (not transparent), requires huge amount of training data and computing power (hours of training using GPUs), is offline and its online versions based on reinforcement learning has no proven convergence, does not guarantee same result for the same input (lacks repeatability) and, more importantly, it does not provide insight, transparency (is a “black-box” type).
In this talk a new, recently introduced approach will be presented which offers highly efficient classifiers, predictive models, etc. but is fully interpretable, transparent and human-intelligible. Moreover, the local optimality as well as the convergence (and respectively, stability) of the proposed systems was theoretically proven and illustrated with examples. The proposed method is prototype-based and non-iterative. It is based on the density and thus it is computationally very efficient (learning on a large amount of images takes few seconds and does not require GPUs or other accelerators as the mainstream deep learning does. However, the performance of the proposed method is on par or better than the competitive alternatives.
The major advantages of this new paradigm is the liberation from the restrictive and often unrealistic assumptions and requirements concerning the nature of the data (random, deterministic, fuzzy), the need to formulate and assume a priori the type of distribution models, membership functions, the independence of the individual data observations, their large (theoretically infinite) number, etc. From a pragmatic point of view, this direct approach from data (streams) to complex, layered model representation is automated fully and leads to very efficient model structures. In addition, the proposed new concept learns in a way similar to the way people learn – it can start from a single example. Thus, the proposed approach is in its nature anthropomorphic.

Short Biography: Prof. Plamen P. Angelov (MEng 1989, PhD 1993, DSc 2015) is a Fellow of the IEEE and of the IET. He is Vice President of the International Neural Networks Society (INNS) and IEEE Distinguished Lecturer. He has 30 years of professional experience in high level research and holds a Personal Chair in Intelligent Systems at Lancaster University, UK and the Director of the LIRA (Lancaster Intelligent, Robotic and Autonomous systems) Research Centre, www.lancaster.ac.uk/lira which is cross-Faculty and has over 30 academics and large number of postdocs and postgraduate students. Previously, he formed and led two research groups (Intelligent Systems, 2010-2013 and Data Science, 2014-2017) at the School of Computing and Communications with over 20 academics, researchers and PhD students each. He has authored or co- authored over 300 peer-reviewed publications in leading journals, peer-reviewed conference proceedings, 6 patents, 3 research monographs (by Wiley, 2012 and Springer, 2002 and 2018) cited over 7700 times with an h-index of 42 and i10-index of 132. He has an active research portfolio in the area of computational intelligence and machine learning and internationally recognised results into online and evolving learning and algorithms for knowledge extraction in the form of human-intelligible fuzzy rule-based systems. Prof. Angelov leads numerous projects (including several multimillion ones) funded by UK research councils, EU, industry, UK MoD. His research was recognised by ‘The Engineer Innovation and Technology 2008 Special Award’ and ‘For outstanding Services’ (2013) by IEEE and INNS. He is also the founding co-Editor-in-Chief of Springer’s journal on Evolving Systems and Associate Editor of several leading international scientific journals, including IEEE Transactions on Cybernetics. He gave over 20 key note talks at high profile conferences. Prof. Angelov was General co-Chair of a number of high profile conferences and a series of annual IEEE Symposia on Evolving and Adaptive Intelligent Systems and more recently on Deep Learning. Dr Angelov is the founding Chair of the Technical Committee on Evolving Intelligent Systems, SMC Society of the IEEE and was previously chairing the Standards Committee of the Computational Intelligent Society of the IEEE (2010-2012). He was also a member of International Program Committee of over 100 international conferences (primarily IEEE). More details can be found at www.lancs.ac.uk/staff/angelov

Prof. Virginio Cantoni

University of Pavia, Italy

Title: Technological innovation and its enhancement of cultural heritage

Abstract: The technological revolution which has completely transformed social relations and which has enabled communication and sharing of multimedia formats is also rapidly transforming the field of art and cultural heritage management. This contribution focuses on the innovative use of interactive digital technologies in digital humanities practices. Multimodal interaction modalities have been developed for the exhibition “1525-2015. Pavia, the Battle, the Future. Nothing was the same again”. Visitors could observe and analyze seven ancient tapestries through 3D reconstructions, virtual simulations, eye interaction and gesture navigation, along with transpositions of the tapestries into tactile images that enable the exploration by partially sighted and blind people. Fired by this successful exhibition, a more ambitious project, the 3D reconstruction of Renaissance Pavia, was undertaken. Advanced techniques and innovative applications have led to a resource that has promoted the history of the city and its architectural richness through videos with virtual tours of Renaissance Pavia. In conclusion, an international challenge, namely “Digital Anastylosis of Frescoes (DAF)”, is proposed: the goal is to Develop learning tools that enable digital reconstruction of destroyed frescoes, supporting their restoration, through the solution of ‘puzzles’ formed by their original fragments, often mixed with spurious elements.

Short Biography: Virginio Cantoni is a Full Professor in Computer Engineering at the University of Pavia. In the period 2008-2011 he has been seconded to the Centro Linceo ‘Beniamino Segre’' of the Italian Academy of Lincei. He has been the founder and first Director of the University of Pavia’s European School of Advanced Studies in Media Science and Technology and Director of the Interdepartmental Centre for Cognitive Science. His research activity is concerned with pattern recognition, computer vision and multimedia. He is author or co-author of more than 300 journal, conference papers and book chapters as well as editor or co-editor of more than 30 books and co-author of five books. He has organized many International Conferences, Seminars and Workshops including a NATO Advanced Research Workshop on pyramidal systems for computer vision. An Expert and Project Reviewer for the EU Commission, he became a Fellow of the IAPR in 1994 and Fellow of the IEEE in 1997.

Prof. Yannis Manolopoulos

Open University of Cyprus, Cyprus

Title: Where to? State-of-the-art POI Recommendations

Abstract: In the recent years, the representation of real-life problems into k-partite graphs introduced a new era in Machine Learning. The combination of virtual and physical layers through Location Based Social Networks (LBSNs) offered a different meaning into the constructed graphs. To this point, multiple diverse models have been introduced in the literature that aim to support users with personalized recommendations. These approaches represent the mathematical models that aim to understand users’ behavior by detecting patterns in users’ check-ins, reviews, ratings and friendships. In this talk, we discuss about state-of-the-art methods for POI recommendations. First, we categorize these methods based on data factors or features they use, the data representation, the methodologies applied and the recommendation types they support. Then, we highlight the most significant limitations and finally we present research trends and future research directions.

Short Biography: Yannis Manolopoulos is Professor and Vice-rector of the Open University of Cyprus. He has been with the University of Toronto, the University of Maryland at College Park, the University of Cyprus and Aristotle University of Thessaloniki, where he served as Head of the Department of Informatics. He has also served as Rector of the University of Western Macedonia in Greece and Vice-Chair of the Greek Computer Society. His research interest focuses in Data Management. He has co-authored 5 monographs and 8 textbooks in Greek, as well as >300 journal and conference papers. He has received >13000 citations from >2000 distinct academic institutions (h-index=52). He has also received 5 best paper awards from SIGMOD, ECML/PKDD, MEDES (2) and ISSPIT conferences and has been invited as keynote speaker in 15 international events. He has served as main co-organizer of several major conferences (among others): ADBIS 2002, SSTD 2003, SSDBM 2004, ICEIS 2006, EANN 2007, ICANN 2010, AIAI 2012, WISE 2013, CAISE 2014, MEDI 2015, ICCCI 2016, TPDL 2017, DAMDID 2017, DASFAA 2018, EAIS 2018, WIMS 2018, IDEAS 2019, MEDES 2019. He has also acted as evaluator for funding agencies in Austria, Canada, Cyprus, Czech Republic, Estonia, EU, Georgia, Greece, Hong-Kong, Israel, Italy, Lithuania, Poland and Russia. Currently, he serves in the Editorial Boards of the following journals (among others): Information Systems, World Wide Web, Computer Journal.

Prof. Ireneusz Czarnowski

Gdynia Maritime University, Poland

Title: Machine Learning and Multiagent Systems

Abstract: In machine learning literature numerous applications solving particular problem type or task, where agent technologies have been played important. Although rather complementary solutions from both fields are discussed the focus is on using agent technology in the field of machine learning with a particular interest on applying agent-based solutions to supervised learning. Some applications are also made with respect to applying machine learning solutions to support agent learning. Based on the research results, where integrating machine learning and agent technologies have been implemented for solving a particular machine learning problem, it can be concluded that such integration strongly support the process of learning capabilities. In my talk I will provide an overview of selected example machine learning models and solutions where the asynchronous team of agents paradigm has been implemented and which have been developed during my researches.

Short Biography: Ireneusz Czarnowski is a graduate of the Faculty of Electrical Engineering at Gdynia Maritime University in 1997. He gained the doctoral degree in the field of computer science in 2004 at Faculty of Computer Science and Management of Poznan University of Technology. In 2012, he earned a postdoctoral degree in the field of computer science in technical sciences at Wroclaw University of Science and Technology. Since 1998 is associated with Gdynia Maritime University, currently is an professor of computer science in the Department of Information Systems at Faculty of Entrepreneurship and Quality Science. Now, he is a Vice Rector for Scientific Research and Development at Gdynia Maritime University.
His main research interests are related to the use of artificial intelligence methods in decision support systems and data mining. His research activity is concerned with machine learning, data reduction for machine learning and data mining. His scientific work also focuses on research in the field of optimization methods and applications of agent systems. Within the field of his interest, he has published, as the author and co-author, more than 100 scientific works, including 15 which appeared in ISI Master Journal List, chapters in scientific monographs, three monographs and several other publications. He has been also the leader of ten research projects and took part in research in five other projects. He is also an editorial board member of the Journals of Knowledge-Based and Intelligent Engineering Systems as well as co-editor several special issues of international journals.
He has been serving as chair or co-chair of programme committee of above 10 international conferences, and served as PC member on more than 40 conferences during the last 3 years. He was co-organizer of several International Conferences and has been co-editor conference proceedings published by Springer and IEEE.
He is also a member of IEEE and, since 2016, he is the Secretary of the Polish branch of IEEE System, Man and Cybernetics and a member of other scientific associations. More details can be found at http://www.umg.edu.pl