On the Fisher Metric of Conditional Probability Polytopes
Guido Montúfar, Johannes Rauh, and Nihat Ay
Contact the author: Please use for correspondence this email.
Submission date: 19. Aug. 2014
published in: Entropy, 16 (2014) 6, p. 3207-3233
DOI number (of the published article): 10.3390/e16063207
Keywords and phrases: Fisher information metric, information geometry, convex support polytope, conditional model, Markov morphism, isometric embedding, natural gradient
Download full preprint: PDF (434 kB)
We consider three different approaches to define natural Riemannian metrics on polytopes of stochastic matrices. First, we define a natural class of stochastic maps between these polytopes and give a metric characterization of Chentsov type in terms of invariance with respect to these maps. Second, we consider the Fisher metric defined on arbitrary polytopes through their embeddings as exponential families in the probability simplex. We show that these metrics can also be characterized by an invariance principle with respect to morphisms of exponential families. Third, we consider the Fisher metric resulting from embedding the polytope of stochastic matrices in a simplex of joint distributions by specifying a marginal distribution. All three approaches result in slight variations of products of Fisher metrics. This is consistent with the nature of polytopes of stochastic matrices, which are Cartesian products of probability simplices. The first approach yields a scaled product of Fisher metrics; the second, a product of Fisher metrics; and the third, a product of Fisher metrics scaled by the marginal distribution.