en
0.25
0.5
0.75
1.25
1.5
1.75
2
Steppest descent analysis for unregularized linear prediction with strictly convex penalties
Published on Jan 25, 20124422 Views
This manuscript presents a convergence analysis, generalized from a study of boosting, of unregularized linear prediction. Here the empirical risk — incorporating strictly convex penalties composed
Related categories
Chapter list
Steepest Descent Analysis for Unregularized Linear Prediction with Strictly Convex Penalties00:00
Setup00:48
Difficulties01:46
Outline - 0103:52
Dual problem04:24
Dual potential function05:26
Upper bounding dual distance06:06
Template theorem10:39
Outline - 0212:45
Minimizers12:50
Unattainability in Boosting I17:08
Unattainability in Boosting II18:43
Summary - 0119:09
Summary - 0219:18