Dr. Pantelis Frangoudis & Dr. Ilir Murturi (TUW) visited UNIZG-FER between February 26th and February 28th, 2024. On November 8th, Dr. Pantelis Frangoudis gave a public talk at UNIZG-FER, hosted by the Croatian Section of IEEE, which was, among other participants, attended by the members of the IoTLab involved in the AIoTwin project. In the mornings of the following two days, Dr. Pantelis Frangoudis and Dr. Ilir Murturi gave two interactive lectures to the members of the IoTLab, which were also streamed online to the rest of the AIoTwin consortium. In the afternoons of those two days, they participated in follow-up research meetings with the members of the IoTLab.
Invited talk: “SMTaaS: Serving problem solving workloads over the computing continuum”
Dr. Pantelis Frangoudis
This invited talk explored how various properties crucial for ensuring the safe, correct, and efficient operation of Cyber-Physical Systems (CPS) can be expressed and verified. An architecture was presented to enable distributed SMT problem solving, addressing significant computational challenges. This framework allowed for dynamic task offloading, optimizing both latency and energy efficiency. Various reinforcement learning-based decision-making strategies were evaluated for their effectiveness in addressing latency minimization and energy efficiency goals. The practicality of the approach was demonstrated through a fog robotics proof-of-concept.
Lecture: “Distributed Computing Continuum Systems: Introduction, technologies and orchestration aspects”
Dr. Pantelis Frangoudis
This presentation highlighted the key challenges in orchestrating Distributed Computing Continuum Systems (DCCS) from both the infrastructure and service provider perspectives. Connectivity technologies for DCCS were also discussed, with a focus on low-power wide area networks and network slicing in 5G.
Lecture: “Distributed Computing Continuum Systems: Emerging approaches for intelligent, self-adaptive and secure operation”
Dr. Ilir Murturi
On the second day, Ilir Murturi and Pantelis A. Frangoudis continued the discussion on DCCS, focusing on "Emerging Approaches for Intelligent, Self-Adaptive, and Secure Operation". Pantelis discussed two dimensions of edge intelligence: AI for the edge and AI on the edge, while introducing an approach by DSG to self-adaptive DCCS utilizing Markov blankets and active inference. Ilir then concluded the lecture by presenting strategies used in DSG for achieving zero trust in DCCS through innovative techniques such as representation learning, blockchain, and federated learning.
.jpg)
.jpg)
.jpg)
Bio: Dr. Pantelis A. Frangoudis is a postdoctoral researcher and university assistant at the Distributed Systems Group, Technical University of Vienna. He received his B.Sc. degree in informatics and the M.Sc. and Ph.D. degrees in computer science from the Department of Informatics, AUEB, Greece, in 2003, 2005, and 2012, respectively, with a thesis on user-centrism in wireless networking. He joined the Distributed Systems Group, TU Wien, in September 2019. Before that, he was a Researcher with the Communication Systems Department, EURECOM, Sophia Antipolis, France, from March 2017 to August 2019, and the Team DIONYSOS, IRISA/INRIA, Rennes, France, from October 2012 to February 2017, which he originally joined under an ERCIM “Alain Bensoussan” Postdoctoral Fellowship. His research interests include mobile/wireless networking, edge computing, network softwarization, network security, cloud computing, content delivery networks, and internet multimedia.
Dr. Ilir Murturi is a Postdoctoral Researcher in the Distributed Systems Group, TU Wien, Austria. In 2022, he obtained his Ph.D. in the Distributed Systems Group, Technische Universität Wien (TU Wien), Vienna, Austria. He received his MSc in Computer Engineering from the University of Prishtina, Kosova. He is an IEEE Member, Guest Editor of the International Journal of Network Management (Wiley), and a reviewer for several IEEE and ACM journals and conferences. Dr. Murturi's research focuses on Distributed Computing Continuum Systems (DCCS), EdgeAI, and privacy in distributed, self-adaptive, and cyber-physical systems. Dr. Murturi's research is partially funded by the AIoTwin project.