BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Reinforcement Learning of Ordinal User Preferences on Wearable Dev
 ices: An Industrial Perspective
DTSTART:20250220T113000Z
DTEND:20250220T123000Z
DTSTAMP:20260316T080900Z
UID:indico-event-72@conferences.enbis.org
DESCRIPTION:Speakers: Davide Cacciarelli (Imperial College London)\n\n\nRe
 inforcement Learning of Ordinal User Preferences on Wearable Devices: An I
 ndustrial Perspective\nSpeaker: Simon Weinberger (ERIC laboratory\, Univ.
  Lyon 2\, France & Essilor International)\nChair: Davide Cacciarelli (Imp
 erial College London\, UK)\nDate: 20th February 2025\, at 12:30-13:30 CET
 \nAbstract:\nFrom personalized treatment plans for patients to robotics an
 d automation and even smart grid optimization\, Reinforcement Learning (RL
 ) has proven effective in numerous problems thanks to its capacity to 
 address complicated sequential decision-making problems.  \nIndeed\, unl
 ike supervised learning\, RL allows addressing dynamic environments where 
 one must take actions depending on some state value in such a way that rew
 ards are maximized over the long run. In the RL paradigm\, taken actions i
 nfluence future states and future rewards\, and this is accounted for\, al
 lowing to approach complex real-world problems where data is unlabeled. \
 nFor example\, wearable devices allow collecting data at an individual lev
 el\, which can be used to propose an unseen degree of personalization for 
 a broad domain of applications\, such as the adaptive automatic control of
  a setting. This can be done using RL: depending on sensor data (state) so
 me setting of the wearable device must be automatically piloted (action) i
 n such a way that the user does not interact often with its device (reward
 ).  \n \nIn this ENBIS webinar\, we provide a brief introduction to Rei
 nforcement Learning and some of its applications and illustrate its use fo
 r the adaptive control of the tint of electrochromic eyewear. With this ob
 jective\, we use an ordinal regression model to control the tint level usi
 ng environmental information from sensors in the eyewear. The time-depende
 nt nature of the predictor is accounted for by considering the state space
  as a functional data space and adapting the Trust Region Policy Optimizat
 ion (TRPO) update accordingly. \nBio:\nSimon Weinberger is a PhD student 
 at ERIC laboratory\, Univ. Lyon 2\, France\, currently working at the Conn
 ected Devices department of EssilorLuxottica. His current role involves ad
 dressing learning strategies for complex wearables devices\, which has bro
 ught him working on Reinforcement Learning and functional data. Since he c
 ompleted a master's degree in applied mathematics\, Simon has been interes
 ted in the use of statistics to solve a wide range of industrial problems.
  \n \n\n\nhttps://conferences.enbis.org/event/72/
URL:https://conferences.enbis.org/event/72/
END:VEVENT
END:VCALENDAR