BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Machine Learning Reading Group @ CUED
SUMMARY:Meta-reinforcement learning - Kris Jensen and Calv
 in Kao (University of Cambridge)
DTSTART;TZID=Europe/London:20210113T110000
DTEND;TZID=Europe/London:20210113T123000
UID:TALK153910AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/153910
DESCRIPTION:Meta learning allows for generalisation across tas
 ks and has become\nincreasingly relevant as machin
 e learning systems are asked to solve\nheterogeneo
 us problems efficiently with less training data. I
 n recent\nyears\, meta learning has been applied i
 n the context of reinforcement\nlearning to build 
 agents that learn to generalise across a distribut
 ion\nof Markov decision problems. In this reading 
 group\, we will briefly\nintroduce the basics of m
 eta reinforcement learning\, cover different\nappr
 oaches to the problem\, and discuss their uses and
  limitations. We\nwill also consider how they comp
 are to more traditional algorithms\, both\nlearned
  and hand-crafted.\n\nRecommended reading:\n\n- Wa
 ng et al. 2016 (https://arxiv.org/abs/1611.05763) 
 OR Duan et al. 2016\n(https://arxiv.org/abs/1611.0
 2779).\n\n- Finn et al. 2017 (https://arxiv.org/ab
 s/1703.03400).\nNagabandi et al. 2019 (https://arx
 iv.org/abs/1803.11347).\n\n- This blog post also p
 rovides an overview of several of the topics we\nw
 ill cover: https://lilianweng.github.io/lil-log/20
 19/06/23/meta-reinforcement-learning.html
LOCATION:https://eng-cam.zoom.us/j/86068703738?pwd=YnFleXFQ
 OE1qR1h6Vmtwbno0LzFHdz09
CONTACT:Elre Oldewage
END:VEVENT
END:VCALENDAR