Dr. Jernej Hribar

is a Marie-Curie Postdoctoral Fellow in the Department of Communication Systems. Previously, he was a Postdoctoral Fellow at Trinity College Dublin (TCD), Ireland, where he served as a working mentor to doctoral and master’s students (2019–2022). He was also a short-term JSPS postdoctoral fellow at Shibaura Institute of Technology, Tokyo, Japan. In 2019, he received his PhD from TCD. His research interests include Age of Information, Deep Reinforcement learning, and Federated Learning.


Research programme: Communication networks and services
Training topic: Managing Smart Infrastructure With Deep Learning

With the world undergoing a digital and green transition, our daily routines increasingly rely on accurate and available information, inspiring us to demand timely and meaningful data while operating in an energy-efficient manner. These challenges are particularly notable in managing complex and interconnected smart infrastructure systems such as smart cities, smart grids, green buildings, etc., where the objective is not only to develop efficient and effective solutions, but also sustainable and environmentally friendly ones. Currently, Artificial Intelligence (AI) is the most promising approach for developing effective management solutions in this context, as it is capable of processing huge amounts of data, recognising patterns, and making predictions while ensuring energy efficiency and environmental sustainability.

In the process of designing the AI-enabled management system, the student will gain hands-on experience in the application of Deep Reinforcement Learning (DRL) algorithms for managing smart infrastructure systems in real time. The student will develop and improve state-of-the-art algorithms such as Deep Deterministic Policy Gradient (DDPG) and learn to design, fine-tune, and evaluate them to optimise performance. Being dependent on real-time provision and processing of time-series data, the research in smart infrastructures also requires a deep understanding of mobile and wireless communication systems and modern techniques for managing radio and network resources in wireless communication networks. The candidate will also apply various algorithms for regression and classification of time series and methods for implementing these algorithms in real-world systems. Working methods will include analytical approaches, computer analysis, and development of deep learning models that will be implemented on an appropriate development platform for testing purposes.

We expect the student to become an independent researcher, whose career will be shaped by the demands and needs of both academia and industry. Special emphasis will be placed on gaining various international experiences, in the form of internships and short-term research visits as well as an opportunity to contribute to EU funded research projects carried out in the research group. As part of PhD research, the student is expected to present their work at highly selective international conferences and publish in prestigious international journals.