Robust approachability and regret minimization in games with partial monitoring thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Robust approachability and regret minimization in games with partial monitoring

Published on Aug 02, 20113088 Views

Approachability has become a standard tool in analyzing learning algorithms in the adversarial online learning setup. We develop a variant of approachability for games where there is ambiguity in th

Related categories

Chapter list

Robust approachability regret minimization with partial monitoring00:00
Introduction00:14
Model01:00
Blackwell’s condition02:44
Sufficiency, insights - 103:52
Sufficiency, insights - 204:25
Sufficiency, insights - 304:39
Sufficiency, insights - 405:30
Uncertainties and partial monitoring06:07
Blackwell strategy to the farthest point - 108:10
Blackwell strategy to the farthest point - 208:42
Blackwell strategy to the farthest point - 308:53
Blackwell strategy to the farthest point - 409:02
Blackwell strategy to the farthest point - 509:26
Blackwell strategy to the farthest point - 609:48
Blackwell strategy to the farthest point - 710:07
Main insights of the result - 110:30
Main insights of the result - 211:37
Main insights of the result - 311:49
Main insights of the result - 411:55
Main insights of the result - 511:59
Main insights of the result - 612:22
Main insights of the result - 713:20
Main insights of the result - 814:04
Main insights of the result - 914:11
Main insights of the result - 1014:12
Main insights of the result - 1114:14
Reduction to full monitoring14:42
[M.,P.,S.] Rates of convergence16:30
Regret with partial monitoring17:57