en
0.25
0.5
0.75
1.25
1.5
1.75
2
Bounds on individual risk for log-loss predictors
Published on Aug 02, 20113148 Views
In sequential prediction with log-loss as well as density estimation with risk measured by KL divergence, one is often interested in the expected instantaneous loss, or, equivalently, the individual r
Related categories
Chapter list
Bounds on Individual Risk for Log-Loss Predictors - 100:00
Bounds on Individual Risk for Log-Loss Predictors - 200:05
Standing in Two Worlds00:17
Standing in Three Worlds - 100:58
Standing in Three Worlds - 201:27
Statistical (Individual) Risk - 102:26
Statistical (Individual) Risk - 203:48
Statistical (Individual) Risk - 304:36
KL Risk and Redundancy - 105:03
KL Risk and Redundancy - 205:32
KL Risk and Redundancy - 306:21
KL Risk and Redundancy - 406:30
Our Question, Informally - 107:18
Our Question, Informally - 207:33
Our Question, Informally - 307:53
Individual Risk of Bayes, Countable M - 109:06
Individual Risk of Bayes, Countable M - 210:25
Individual Risk of Bayes, Countable M - 312:06
Parametric Case - 112:50
Parametric Case - 214:02
Parametric Case - 314:06
Parametric Case: Taylor no good14:29
Parametric Case-Decreasing Risk - 115:02
Parametric Case-Decreasing Risk - 215:20
Problem: Bayes is not “smart”16:13
Related Fundamental Question17:10
Final Remarks17:58