Updating Probabilities

Abstract

We show that Skilling's method of induction leads to a unique general theory of inductive inference, the method of Maximum relative Entropy (ME). The main tool for updating probabilities is the logarithmic relative entropy; other entropies such as those of Renyi or Tsallis are ruled out. We also show that Bayes updating is a special case of ME updating and thus, that the two are completely compatible.

Presented at MaxEnt 2006, the 26th International Workshop on Bayesian Inference and Maximum Entropy Methods (July 8-13, 2006, Paris, France)

Cited by in corpus (12)

• Jacobi Fields on Statistical Manifolds of Negative Curvature
• Fully Flexible Views: Theory and Practice
• Information-Geometric Indicators of Chaos in Gaussian Models on Statistical Manifolds of Negative Ricci Curvature
• Using Relative Entropy to Find Optimal Approximations: an Application to Simple Fluids
• Information Geometry and Chaos on Negatively Curved Statistical Manifolds
• Geometrodynamics of Information on Curved Statistical Manifolds and its Applications to Chaos
• Generalized Galilean Transformations and the Measurement Problem in the Entropic Dynamics Approach to Quantum Theory
• Maximum Entropy: The Universal Method for Inference
• Unification of some classical and quantum ideas
• Updating Probabilities: An Econometric Example
• The relativity of theory
• Reexamination of an Information Geometric Construction of Entropic Indicators of Complexity