Sign inSign up
Learning to Communicate with Deep Multi­-Agent Reinforcement Learning thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Learning to Communicate with Deep Multi­-Agent Reinforcement Learning

Published on Aug 23, 20167899 Views

Jakob Foerster

Deep Learning Summer School 2016 - Montreal

Related categories

Deep LearningReinforcement LearningUnsupervised Learning

Chapter list

Learning to Communicate with Deep Multi-Agent RL00:00
Motivation00:54
Background and Setting01:36
Methods - 103:10
Methods - 203:49
Methods - DRU06:16
Methods - Architecture06:49
Experiments - Switch07:20
Experiments - Switch Complexity Analysis08:46
Experiments - Switch09:11
Experiments - Switch Strategy10:12
Experiments - MNIST Games10:59
Experiments - MNIST Game11:00
Experiments - MNIST Results and Strategy11:40
Experiments - Impact of Noise12:22
Future Work13:06
Conclusions13:32
Thank you14:33
Q&A14:36