BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning in Feedback Loops: Cooperation and Performativity in Mult
 i-Agent Systems - Stefanos  Leonardos (King's College London)
DTSTART:20251112T094000Z
DTEND:20251112T102000Z
UID:TALK238486@talks.cam.ac.uk
DESCRIPTION:This talk explores how feedback between learning agents and th
 eir environment shapes collective behaviour and market dynamics. In the fi
 rst part\, I present a decentralized multi-agent reinforcement learning fr
 amework where agents develop cooperation through participation incentives\
 , even without explicit coordination or shared information. In the second 
 part\, I discuss performative market making&mdash\;how financial models ca
 n become self-fulfilling by influencing the very markets they aim to descr
 ibe. Together\, these studies reveal how stochastic control and reinforcem
 ent learning intertwine in systems where learning agents not only adapt to
 \, but also create\, their environment.\nCo-authors: Stefan Roesch (PhD st
 udent at KCL)\, Prof. Odinaldo Rodrigues (KCL)\, Dr. Yali Du (KCL)\, Chara
 lampos Kleitsikas (PhD student at KCL)\, Prof. Carmine Ventre (KCL).
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR