A channel-based perspective on conjugate priors

DOI10.1017/S0960129519000082zbMATH Open1442.62054OpenAlexW3007521435MaRDI QIDQ5220182FDOQ5220182

Authors:

Publication date: 11 March 2020

Published in: Mathematical Structures in Computer Science (Search for Journal in Brave)

Abstract: A desired closure property in Bayesian probability is that an updated posterior distribution be in the same class of distributions --- say Gaussians --- as the prior distribution. When the updating takes place via a statistical model, one calls the class of prior distributions the `conjugate priors' of the model. This paper gives (1) an abstract formulation of this notion of conjugate prior, using channels, in a graphical language, (2) a simple abstract proof that such conjugate priors yield Bayesian inversions, and (3) a logical description of conjugate priors that highlights the required closure of the priors under updating. The theory is illustrated with several standard examples, also covering multiple updating.

Full work available at URL: https://arxiv.org/abs/1707.00269

Recommendations

• Conjugacy of probability distributions in statistical problems
• Extensions of the conjugate prior through the Kullback--Leibler separators
• scientific article; zbMATH DE number 370368
• Disintegration and Bayesian inversion via string diagrams
• Conjugate Exponential Family Priors For Exponential Family Likelihoods

zbMATH Keywords

conjugate priorposterior distributionBayesian inversionBayesian probability

Mathematics Subject Classification ID

Bayesian inference (62F15) Censored data models (62N01)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• A categorical foundation for Bayesian probability
• A survey of graphical languages for monoidal categories
• An effect-theoretic account of Lebesgue integration
• Categorical quantum mechanics
• Conjugate priors for exponential families
• Dagger compact closed categories and completely positive maps (extended abstract)
• From probability monads to commutative effectuses
• Introduction to machine learning.
• Labelled Markov processes.
• Measurable spaces and their effect logic
• New directions in categorical logic, for classical, probabilistic and quantum logic
• Pattern recognition and machine learning.
• Pointless learning
• The Logical Essentials of Bayesian Reasoning
• The existence of regular conditional probabilities: Necessary and sufficient conditions

Cited In (4)

• The general expression of the prior convergence error: a proof
• A non-commutative Bayes' theorem
• Multisets and distributions, in drawing and learning
• A geometric view of conjugate priors

Uses Software

• PRMLT

This page was built for publication: A channel-based perspective on conjugate priors

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5220182)