en
0.25
0.5
0.75
1.25
1.5
1.75
2
Bandits, Query Learning, and the Haystack Dimension
Published on Aug 02, 20112894 Views
Motivated by multi-armed bandits (MAB) problems with a very large or even in finite number of arms, we consider the problem of finding a maximum of an unknown target function by querying the function
Chapter list
Bandits, Query Learning, and the Haystack Dimension00:00
Motivation00:03
Problem Formulation00:47
Main Results, Pictorially - 101:41
Main Results, Pictorially - 202:15
Maximizing from Queries02:29
Haystack Dimension Intuition03:32
More Haystack Dimension Intuition - 104:29
More Haystack Dimension Intuition - 204:35
More Haystack Dimension Intuition - 304:45
More Haystack Dimension Intuition - 404:53
More Haystack Dimension Intuition - 504:57
More Haystack Dimension Intuition - 605:08
The Haystack Dimension, Precisely - 105:24
The Haystack Dimension, Precisely - 205:53
The Haystack Dimension, Precisely - 306:49
The Haystack Dimension, Precisely - 406:57
The Haystack Dimension, Precisely - 507:08
The Haystack Dimension, Precisely - 607:27
The Haystack Dimension, Precisely - 707:33
The Haystack Dimension, Precisely - 807:43
The Haystack Dimension, Precisely - 907:45
Lower Bound for MAX Problem08:38
Lower Bound Proof Sketch - 109:16
Lower Bound Proof Sketch - 209:54
Lower Bound Proof Sketch - 310:13
Lower Bound Proof Sketch - 410:22
Lower Bound Proof Sketch - 510:40
Lower Bound Proof Sketch - 611:04
Greedy MAX11:46
Returning to MAB Setting13:45
Infinite F15:21
Conclusions16:02
Thanks!16:36