BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Discrete-time total stability: learning robust flux-profile contro
 llers for tokamak reactors
DTSTART:20230303T110000
DTEND:20230303T120000
DTSTAMP:20260410T010031Z
UID:228190bd48c8dac2f52f0a5d8df1559ed8d077dbdddec3ce1bab5851
CATEGORIES:Conferences - Seminars
DESCRIPTION:Samuele Zoboli \nAbstract: Recent advances in Artificial Inte
 lligence-based control provide a suitable framework for plasma profile con
 trol. However\, most of these learning algorithms lack fundamental robustn
 ess and generalization guarantees. This impairs their applicability outsid
 e of simulation. In this talk\, we present novel total stability results f
 or discrete-time nonlinear systems\, allowing the study of stability prope
 rties via model comparison. These theoretical foundations justify the use 
 of Deep Reinforcement Learning algorithms to train controllers in a simula
 ted environment\, and their subsequent application in real-world scenarios
 . We present experimental results showing the effectiveness of such learne
 d agents equipped with an error integrator in robustly controlling the saf
 ety factor profile in Tokamak reactors. \n\nBio: Samuele Zoboli received 
 the B.S. degree in electronics engineering from the University of Modena a
 nd Reggio Emilia\, Italy\, in 2016 and the M.S degree with honors in autom
 ation engineering from University of Bologna\, Italy\, in 2019. He current
 ly is a Ph.D. candidate at LAGEPP\, University of Lyon 1\, France. His res
 earch interests include stabilization of discrete-time nonlinear systems\,
  multi-agent systems\, reinforcement learning and control-applied artifici
 al intelligence.
LOCATION:ME C2 405 https://plan.epfl.ch/?room==ME%20C2%20405
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR