BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Towards a single controller that can make many buildings a little 
 greener - Arduin Findeis
DTSTART:20221014T120000Z
DTEND:20221014T130000Z
UID:TALK183257@talks.cam.ac.uk
CONTACT:Madeline Lisaius
DESCRIPTION:"The operation of buildings is responsible for approximately 2
 8% of all energy-related CO2 emissions. Despite causing such a large propo
 rtion of emissions\, most buildings are not operated as well as they could
  be\, resulting in an untapped potential to reduce emissions. Recognising 
 this opportunity\, there has been extensive work on improving building con
 trollers using advanced control methods\, such as model predictive control
  (MPC) or reinforcement learning (RL). These state-of-the-art methods have
  been proven to reduce emissions on test systems. Despite these results\, 
 few of these controllers are used outside research settings as most method
 s require prohibitively expensive expert tuning to be applied to any new b
 uilding.\n\nTo make it easier to deploy more advanced building controllers
 \, we are working on generalist building controllers that can control a wi
 de variety of buildings - with the same controller and no further expert t
 uning. These controllers have the potential to make many buildings operate
  a little greener. In this talk\, I will outline how we plan to develop su
 ch generalist controllers and will present preliminary results. In particu
 lar\, I will demo our open-source building simulation framework\, named Ba
 uwerk\, and discuss early results of applying multi-task and meta reinforc
 ement learning (RL) methods to this experimental setup."
LOCATION:Seminar time is 1pm BST in Room FW 11\, Willam Gates Hall. Zoom l
 ink: https://cl-cam-ac-uk.zoom.us/j/4361570789?pwd=Nkl2T3ZLaTZwRm05bzRTOUU
 xY3Q4QT09&amp\;from=addon 
END:VEVENT
END:VCALENDAR