BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Control of complex dynamics using reinforcement learning
DTSTART:20250521T140000
DTEND:20250521T150000
DTSTAMP:20260601T073040Z
UID:d44172791be190026e3e062c4869fe9fa4c27ec095fb7457c0802ab0
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr Onofrio Semeraro\, CNRS Research Associate\nAbstract\n\nIn 
 this talk\, we will focus on flow control -- a subject that has attracted 
 numerous research efforts in the last decades -- and more specifically Rei
 nforcement Learning (RL) as an alternative framework to standard technique
 s of control. RL encompasses a large variety of algorithms and strategies 
 and has gained traction in recent years\, due to the possibility of tackli
 ng the control of nonlinear systems while circumventing preliminary modell
 ing steps. Indeed\, RL algorithms do not require any a-priori knowledge of
  the equations governing the system to be controlled and solely rely on th
 e local measurements of the flow\, based on which a policy is learnt from 
 the interaction of the agent with the environment. Despite successes\, the
  first documented applications of RL for control in fluid dynamics often r
 esult in highly non-intuitive control policies\, also when cheaper optimal
  solutions are available. We will discuss some examples of fluid mechanics
  interest\, highlighting challenges as well as possible remedies for the c
 urrent practice.\n\n================================\n\nBiography\n\nOnofr
 io Semeraro received his PhD in Mechanical Engineering at KTH-Stockholm (S
 weden) in 2013. He served as postdoctoral researcher at Ecole-Polytechniqu
 e\, Palaiseau (France) and Politecnico of Bari (Italy)\, and he is current
 ly a CNRS Research Associate since 2017\, at the Laboratoire Interdiscipli
 naire des Sciences du Numérique (LISN) - Universite Paris Saclay\, Orsay 
 (France). His studies focus mainly on control\, data assimilation\, modell
 ing and data-driven techniques\, ranging from system identification to dee
 p learning for fluid mechanics.
LOCATION:MED 1 1518 https://plan.epfl.ch/?room==MED%201%201518
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR