107 search results - page 4 / 22 » General Loss Bounds for Universal Sequence Prediction |
COLT
1999
Springer
13 years 11 months ago
1999
Springer
We present a unified framework for reasoning about worst-case regret bounds for learning algorithms. This framework is based on the theory of duality of convex functions. It brin...
NIPS
2001
13 years 8 months ago
2001
In this paper, it is shown how to extract a hypothesis with small risk from the ensemble of hypotheses generated by an arbitrary on-line learning algorithm run on an independent an...
NIPS
2008
13 years 8 months ago
2008
Large-margin structured estimation methods minimize a convex upper bound of loss functions. While they allow for efficient optimization algorithms, these convex formulations are n...
JCSS
2006
13 years 7 months ago
2006
This paper studies sequence prediction based on the monotone Kolmogorov complexity Km=-log m, i.e. based on universal deterministic/one-part MDL. m is extremely close to Solomonof...
COLT
2004
Springer
14 years 22 days ago
2004
Springer
We study the properties of the Minimum Description Length principle for sequence prediction, considering a two-part MDL estimator which is chosen from a countable class of models....