| KEYWORD |
Improving Transport and Routing Protocols via Reinforcement Learning
Tesi all'estero
Riferimenti GUIDO MARCHETTO, ALESSIO SACCO
Riferimenti esterni Prof. Flavio Esposito - St. Louis University, USA
Gruppi di ricerca DAUIN - GR-03 - COMPUTER NETWORKS GROUP - NETGROUP
Descrizione The computer networking community has been steadily increasing investigations into machine learning to help solve tasks such as routing, traffic prediction, and resource management. In particular, due to the recent successes in other applications, Reinforcement Learning (RL) has seen steady growth in network management and, more recently, in routing. However, changes in the network topology prevent RL-based routing approaches from being employed in real environments due to the need for retraining. In this project, the student will design algorithms that approach routing as an RL problem with two novel twists: 1) minimizing flow set collisions and 2) dealing with routing in dynamic network conditions without retraining. The student will compare routing protocols, using real routers and simulations, and will learn multi-agent reinforcement learning, and how to evaluate a protocol according to various networks and machine learning metrics sharing our lesson learned in a scientific publication.
Scadenza validita proposta 08/02/2025
PROPONI LA TUA CANDIDATURA