en
0.25
0.5
0.75
1.25
1.5
1.75
2
Contextual Bandits with Similarity Information
Published on Aug 02, 20113729 Views
In a multi-armed bandit (MAB) problem, an online algorithm makes a sequence of choices. In each round it chooses from a time-invariant set of alternatives and receives the payoff associated with thi
Chapter list
Contextual Bandits with Similarity Information00:00
Running example00:00
Multi-Armed Bandits00:18
Contextual Bandits01:12
Similarity info - 102:32
Similarity info - 203:55
Problem formulation04:51
Some background06:23
Prior work: uniform partitions - 107:20
Prior work: uniform partitions - 209:21
This paper: adaptive partitions11:04
Rest of the talk12:22
Algorithm: contextual zooming - 112:49
Algorithm: contextual zooming - 214:10
Unexpected application:slowly changing payoffs16:00
Unexpected application: sleeping bandits17:52
Other results18:53