- Mirjana Ivanović (University of Novi Sad, Serbia).
Title: Influence of Federated Learning on Contemporary Distributed Machine Learning Research
Abstract: Federated Learning (FL) is a modern distributed machine learning approach based on a collaboratively decentralized privacy-preserving technology to overcome challenges of data silos and data sensibility. It supports multiple, wide range of clients (like mobile devices but also different institutions, organizations, etc.) that are coordinated with one or more central servers performing decentralized machine learning.
Although it is rather new approach numerous strategies for implementing and organizing FL have been developing. The efficiency of selected FL strategy for some particular problem and domain is influenced by taking into account the organizational structure and the actors involved in it.
FL plays an essential role in environments and application when high level of data privacy preserving, and security is needed so particular attention should be paid to adoption to these aspects. Possible domains of applications are numerous and includes banking and financial sector, IoT and smart environments, automotive industry, medical and health domains etc..
In this presentation we will focus on crucial aspects of FL paradigm, present its essential characteristics, especially put focus on privacy preserving techniques employed within FL approaches, and present several characteristic examples of applications. Additionally, some promising directions to lead future development and use of FL in different application domains will be briefly introduced.Full Professor at Faculty of Sciences, University of Novi Sad, Serbia
Mirjana Ivanovic holds the position of Full Professor at Faculty of Sciences, University of Novi Sad, Serbia. She is a member of National Scientific Committee for Electronics, Telecommunication and Informatics within Ministry of Education, Science and Technological Development, Republic of Serbia and member of Board of directors of the Institute for Artificial Intelligence Research and Development of Republic of Serbia. She was a member of the University Council for Informatics for more than 12 years. Prof. Ivanovic is the author or co-author of 14 textbooks, several international monographs and more than 450 research papers, most of which are published in international journals and conferences. Her research interests include agent technologies, intelligent techniques, applications of data mining and machine learning techniques in medical domains and technology enhanced learning. She is a member of Program Committees of more than 350 international conferences, Program/General Chair of several international conferences, and leader of numerous international research projects. Mirjana Ivanovic delivered numerous keynote speeches at international conferences and visited many academic institutions all over the world as visiting researcher and teacher (Germany, North Macedonia, Slovenia, Portugal, Australia, China, Korea). Currently she is Editor-in-Chief of the Computer Science and Information Systems journal and Associated Editor in several international Journals. - Florian Dörfler (Swiss Federal Institute of Technology Zürich, Switzerland).
Title: Online Feedback Optimization
Abstract: Online feedback optimization refers to the design of feedback controllers that asymptotically steer a physical system to the solution of an optimization problem while respecting physical and operational constraints. For the considered optimization problem many parameters might be unknown, but one can rely on real-time measurements and the underlying physical system enforcing certain constraints. This problem setup is motivated by applications to electric power systems and has historic roots in communication networks and process control. In comparison to other optimization-based control strategies, transient optimality of trajectories is not the primary goal, and no predictive model, running costs or exogenous set-points are required. Hence, one aims at controllers that require little model information, demand low computational cost, but that leverage real-time measurements. We design such controllers based on optimization algorithms that take the form of open and discontinuous dynamical systems. In this talk we discuss different algorithms such as projected gradient and saddle-point flows, their closed-loop stability when interconnected with physical systems, robustness properties, regularity conditions, and implementation aspects. Throughout the talk we demonstrate the potential of our methodology for real-time operation of power systems including industrial implementations.Florian Dörfler is an Associate Professor at the Automatic Control Laboratory at ETH Zürich and the Associate Head of the Department of Information Technology and Electrical Engineering. He received his Ph.D. degree in Mechanical Engineering from the University of California at Santa Barbara in 2013, and a Diplom degree in Engineering Cybernetics from the University of Stuttgart in 2008. From 2013to 2014 he was an Assistant Professor at the University of California Los Angeles. He is a recipient of the distinguished young research awards by IFAC (Manfred Thoma Medal 2020) and EUCA (EuropeanControl Award 2020). His students were winners or finalists for Best Student Paper awards at the European Control Conference (2013, 2019), American Control Conference (2016), Conference on Decisionand Control (2020), PES General Meeting (2020), and PES PowerTech Conference (2017). He received the 2010 ACC Student Best Paper Award, the 2011 O. Hugo Schuck Best Paper Award, the 2012-2014 Automatica Best Paper Award, the 2016 IEEE Circuits and Systems Best Paper Award, and the 2015UCSB ME Best PhD award.
Florian Dörfler’s research interests are centered around distributed control and optimization in complex, cyber-physical, and networked systems with applications to smart power grids, robotic coordination, and social networks. Topics of recent interest include stability and control in low-inertia power grids, online feedback optimization with applications to power systems operation, distributed and plug-and- play control and optimization, data-driven black-box control, and synchronization in complex networks. - Ming Cao (University of Groningen, The Netherlands).
His research interests are in multi-agent systems, complex systems and networks, distributed decision-making and coordination, cooperative control, adaptive control, human/robot interaction dynamics, robotic teams, brain and neural networks, electric power systems, traffic networks.
The research of Cao has been supported in part by the European Research Council (ERC), the European Union under its Horizaon 2020, Seventh Framework Program and INTERREG program, the Dutch Organization for Scientific Research (NWO), and other international and regional funding agencies.