Skip to content

Entropy and inference, revisited

Ilya Nemenman, Fariel Shafee, William Bialek

physics.data-anarXiv:physics/0108025

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.

Create a lesson