Jacopo Castellini

Bio

After receiving a Bachelor (2014) and a Master (2017) degrees in Computer Science at the Università degli Studi di Perugia, in 2017 I moved to the University of Liverpool for a Ph.D. in machine learning, specifically in multi-agent reinforcement learning. After being granted the Ph.D. in 2022, I obtained a 9 months postdoc position at the Centre for Competition Policy at the University of East Anglia, working on a CMA-funded project on the impact of recommender systems on markets competition and their policy making. Then, in 2023, I started a 8 months postdoc position at INSA Lyon, working on optimal-guaranteed planning in partially observable multi-agent systems. Finally, in 2024, I moved to HES-SO Genève to work on the 3-years EU project Hyper-AI.

Research

I am interested in the field of machine learning, especially in multi-agent reinforcement learning. My current work is on decentralised reinforcement learning for data-related resource management in the edge-cloud continuum, and more broadly in designing novel and better algorithms for decentralized multi-agent reinforcement learning. Other research interests are related to neural networks, deep learning, planning and multi-agent systems.

Publications

A list of publications can be found here.