Strong Asymptotic Assertions for Discrete MDL in Regression and Classification

Details Strong Asymptotic Assertions for Discrete MDL in Regression and Classification Strong Asymptotic Assertions for Discrete MDL in Regression and Classification

2005-02-15

arxiv.org/abs/math/0502315v1

Proc. 14th Dutch-Belgium Conf. on Machine Learning (Benelearn 2005) 67-72

Mathematics

Statistics Theory

6 two-column pages

Scientific paper

We study the properties of the MDL (or maximum penalized complexity) estimator for Regression and Classification, where the underlying model class is countable. We show in particular a finite bound on the Hellinger losses under the only assumption that there is a "true" model contained in the class. This implies almost sure convergence of the predictive distribution to the true one at a fast rate. It corresponds to Solomonoff's central theorem of universal induction, however with a bound that is exponentially larger.

Affiliated with

Also associated with

No associations

Rating

Strong Asymptotic Assertions for Discrete MDL in Regression and Classification does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Strong Asymptotic Assertions for Discrete MDL in Regression and Classification, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Strong Asymptotic Assertions for Discrete MDL in Regression and Classification will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-242855