Horizon-Independent Optimal Prediction with Log-Loss in Exponential Families thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Horizon-Independent Optimal Prediction with Log-Loss in Exponential Families

Published on Aug 09, 20133881 Views

We study online learning under logarithmic loss with regular parametric models. Hedayati and Bartlett (2012) showed that a Bayesian prediction strategy with Jeffreys prior and sequential normalized ma

Related categories

Chapter list

Horizon-Independent Optimal Prediction with Log-Loss in Exponential Families00:00
Player Adversary's Game00:09
Goal00:57
Sequential Probability Assignment Equivalent to Joint Distribution01:47
Normalized maximum likelihood04:23
Bayesian strategies04:53
SNML05:01
Exponential families05:10
Implications05:16
End05:17