0.25
0.5
0.75
1.25
1.5
1.75
2
On-line linear learning algorithms
Published on Feb 25, 20076330 Views
Prediction with expert advice. Learning with linear experts. The Perceptron algorithm and its extensions. On-line learning with kernels. Mistake bounds. From mistake bounds to risk bounds.
Related categories
Chapter list
On-Line Learning00:02
Background00:25
Outline09:09
Binary prediction11:41
The role of experts13:52
A simple example17:24
A more general prediction model20:10
On-line prediction with expert advice23:04
Regret26:38
On-line prediction with expert advice31:06
Regret31:25
Weighted average forecasters31:45
Convex loss functions35:37
Potential-based forecasters37:28
Convex loss functions37:47
Blackwell condition41:22
Gradient descent interpretation42:27
Regret43:41
Gradient descent interpretation43:48
Blackwell condition44:44
Gradient descent interpretation44:57