SCIENCE@home

Universal Batch Learning with Log-Loss

Yaniv Fogel,M. Feder

Published 2018 in International Symposium on Information Theory

ABSTRACT

In this paper we consider the problem of batch learning with log-loss, in a stochastic setting where given the data features, the outcome is generated by an unknown distribution from a class of models. Utilizing the minimax theorem and information-theoretical tools, we came up with the minimax universal learning solution, a redundancy capacity theorem and an upper bound on the performance of the optimal solution. The resulting universal learning solution is a mixture over the models in the considered class. Furthermore, we get a better bound on the generalization error that decays as $O(\log N/N)$, where $N$ is the sample size, instead of $O(\sqrt{\log N/N})$ which is commonly attained in statistical learning theory for the empirical risk minimizer.

PUBLICATION RECORD

CITATION MAP

EXTRACTION MAP

CLAIMS

  • No claims are published for this paper.

CONCEPTS

  • No concepts are published for this paper.

REFERENCES

Showing 1-12 of 12 references · Page 1 of 1

CITED BY

Showing 1-25 of 25 citing papers · Page 1 of 1