AI-assisted optimization of cloud gaming experience
Riferimenti PAOLO GIACCONE
Riferimenti esterni German Sviridov (email@example.com)
Gruppi di ricerca Telecommunication Networks Group
Tipo tesi SIMULATIVE AND EXPERIMENTAL
Descrizione Recent advances in Deep Reinforcement Learning showed how artificial agents are able to play even complex video games (e.g., Doom, Starcraft, Super Mario Smash Bros) at human level. While the research community is focused on developing complex AI with the goal of outperforming human players, possible applications in the field of networking are still to be explored.
This thesis will focus on the analysis of the Quality of Experience (QoE) in video games by considering the emerging cloud gaming scenario.
Artificial agents will be used to emulate human players and the obtained results will be employed to perform game-aware network optimization (e.g., scheduling, resource-allocation, traffic prioritization).
Scadenza validita proposta 20/09/2020 PROPONI LA TUA CANDIDATURA