Algorithmic statistics: normal objects and universal models
Abstract
Kolmogorov suggested to measure quality of a statistical hypothesis for a data by two parameters: Kolmogorov complexity of the hypothesis and the probability of with respect to . P. G\'acs, J. Tromp, P.M.B. Vit\'anyi discovered a small class of models that are universal in the following sense. Each hypothesis from that class is identified by two integer parameters and for every data and for each complexity level there is a hypothesis with of complexity at most that has almost the best fit among all hypotheses of complexity at most . The hypothesis is identified by and the leading bits of the binary representation of the number of strings of complexity at most . On the other hand, the initial data might be completely irrelevant to the the number of strings of complexity at most . Thus seems to have some information irrelevant to the data, which undermines Kolmogorov's approach: the best hypotheses should not have irrelevant information. To restrict the class of hypotheses for a data to those that have only relevant information, Vereshchagin introduced a notion of a strong model for : those are models for whose total conditional complexity conditional to is negligible. An object is called normal if for each complexity level at least one its best fitting model of that complexity is strong. In this paper we show that there are "many types" of normal strings. Our second result states that there is a normal object such that all its best fitting models are not strong for . Our last result states that every best fit strong model for a normal object is again a normal object.
Cite
@article{arxiv.1512.04510,
title = {Algorithmic statistics: normal objects and universal models},
author = {Alexey Milovanov},
journal= {arXiv preprint arXiv:1512.04510},
year = {2015}
}
Comments
19 pages, 5 figures