English

Entropy and inference, revisited

Data Analysis, Statistics and Probability 2007-05-23 v2

Abstract

We study properties of popular near-uniform (Dirichlet) priors for learning undersampled probability distributions on discrete nonmetric spaces and show that they lead to disastrous results. However, an Occam-style phase space argument expands the priors into their infinite mixture and resolves most of the observed problems. This leads to a surprisingly good estimator of entropies of discrete distributions.

Keywords

Cite

@article{arxiv.physics/0108025,
  title  = {Entropy and inference, revisited},
  author = {Ilya Nemenman and Fariel Shafee and William Bialek},
  journal= {arXiv preprint arXiv:physics/0108025},
  year   = {2007}
}

Comments

LaTex2e, 9 pages, 5 figures; references added, minor revisions introduced, formatting errors corrected