0.25
0.5
0.75
1.25
1.5
1.75
2
Applications of bandits and recommendation systems
Published on Jul 27, 20174033 Views
Related categories
Chapter list
RL in the industry00:00
Disclaimer00:16
Two components in RL00:54
Retargeting: how it works02:14
Details of the auction03:35
Finding a bidding strategy05:21
A/B testing06:15
RMSE vs. true revenue06:54
Implicit assumptions08:24
Quality of the proxy09:46
Is the input distribution the same?15:29
Simpson’s paradox16:28
Dealing with confounding variables20:03
Exploring exploration21:31
Misspecified model30:48
Counterfactual question31:03
Counterfactual question - 131:53
Offline vs. online evaluation33:29
From evaluation to optimization34:51
Benefits of policy evaluation36:45
Efficient policy optimization39:18
Other unanswered questions40:43
Executive summary41:45
Thank you!43:40