Entropic SMML defines a one-parameter family of coding rules that interpolates between Bayesian average-case and minimax worst-case codelengths, with a PAC-Bayes variational form and asymptotic regimes for regular parametric models.
Wallace.Statistical and inductive inference by minimum message length
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
math.ST 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
SMML estimators asymptotically induce information-geometric quantisation where partitions are pullbacks of weighted Fisher-Rao Voronoi tessellations and codepoints are KL centroids for exponential families.
citing papers explorer
-
Entropic Strict Minimum Message Length and Its Connections to PAC-Bayes and NML
Entropic SMML defines a one-parameter family of coding rules that interpolates between Bayesian average-case and minimax worst-case codelengths, with a PAC-Bayes variational form and asymptotic regimes for regular parametric models.
-
Information Geometry and Asymptotic Theory for SMML Estimators
SMML estimators asymptotically induce information-geometric quantisation where partitions are pullbacks of weighted Fisher-Rao Voronoi tessellations and codepoints are KL centroids for exponential families.