Sign inSign up
Thompson Sampling for Learning Parameterized Markov Decision Processes thumbnail
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Thompson Sampling for Learning Parameterized Markov Decision Processes

Published on Aug 20, 20151727 Views

Aditya Gopalan

We consider reinforcement learning in parameterized Markov Decision Processes (MDPs), where the parameterization may induce correlation across transition probabilities or rewards. Consequently, obser

COLT 2015 - Paris

Related categories

Active LearningComputational Learning TheoryOn-line LearningReinforcement LearningSemi-supervised Learning

Chapter list

Thompson Sampling for Learning Parameterized Markov Decision Processes00:00
Online Reinforcement Learning - 100:02
Online Reinforcement Learning - 200:15
Online Reinforcement Learning - 300:19
Online Reinforcement Learning - 400:28
Online Reinforcement Learning - 500:34
Online Reinforcement Learning - 600:38
Online Reinforcement Learning - 700:42
Online Reinforcement Learning - 800:44
Online Reinforcement Learning - 900:45
Online Reinforcement Learning - 1000:47
Online Reinforcement Learning - 1100:48
Online Reinforcement Learning - 1200:49
Online Reinforcement Learning - 1300:50
Online Reinforcement Learning - 1400:51
Online Reinforcement Learning - 1500:51
Online Reinforcement Learning - 1600:53
Online Reinforcement Learning - 1700:54
Online Reinforcement Learning - 1800:54
Online Reinforcement Learning - 1900:55
Online Reinforcement Learning - 2000:56
Thompson Sampling [Thompson 1933] - 101:36
Thompson Sampling - 101:57
Thompson Sampling - 202:01
Thompson Sampling - 302:16
Main Result - 102:27
Main Result - 202:53