An optimal visiting mean field game problem

Luogo: l'evento si terrà per via telematica attraverso collegamento alla piattaforma Zoom. Per partecipare sarà necessario collegarsi a questo link e inserire i codici di accesso:

Meeting ID: 455 516 299
Password: 018744

Ore: 15.00 

Relatori:

• Fabio Bagagiolo (Università di Trento)    
• Luciano Marzufero (Università di Trento)

Abstract:

A very large number of agents want to visit a finite number of sites avoiding queues and congested spots as more as possible. The problem is modeled as a mean field game problem where the population of agents is split, at every instant, into several populations labelled by the "string" of the already visited sites. From an analytical point of view, this is realized by a suitable system of Hamilton-Jacobi equations (or quasi-variational inequalities) and transport (conservation laws-like) equations. The seminar will possibly end with some difficulties on the analytical model that we have encountered (and not completely solved yet), but a first introductory part will try to explain the model and its analytical representation, in a possibly rather intuitive way.  

Referente: Fabio Bagagiolo