BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Solving mean-field stochastic control problems by using deep learn
 ing - Nacira Agram (KTH Stockholm)
DTSTART:20220420T103000Z
DTEND:20220420T111500Z
UID:TALK171419@talks.cam.ac.uk
DESCRIPTION:The two famous approaches of solving stochastic control proble
 ms are Bellman&rsquo\;s dynamic programming and Pontryagin&rsquo\;s maximu
 m principle. The dynamic programming method can be very efficient\, but it
  works only if the system is Markov.&nbsp\;The maximum principle\, on the 
 other hand\, does not require that the system is Markov\, but it has the d
 rawback that it involves complicated backward stochastic differential equa
 tions. The mean-field systems are not Markovian a priori\, but they can be
  made Markovian by adding to the system the Fokker-Planck equation for the
  law if the state. &nbsp\;Then we can use the dynamic programming to study
  optimal&nbsp\;control&nbsp\;of of mean-field equations.\nMean-field dynam
 ics have a lot of applications\, in this talk I will represent in particul
 ar two applications: Optimal energy consumption by the cortex neural netwo
 rk and initial investment problems. We will apply stochastic control metho
 ds to solve the problems. Furthermore\, it is sometimes difficult to find 
 explicit solutions mathematically and therefore\, we will use numerical me
 thod to find them.\nWe will use deep learning technics to solve special ca
 ses of the above discussed problems explicitly.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR