Abstract:Divergence functions are the non-symmetric “distance” on the manifold, Μθ, of parametric probability density functions over a measure space, (Χ,μ). Classical information geometry prescribes, on Μθ: (i) a Riemannian metric given by the Fisher information; (ii) a pair of dual connections (giving rise to the family of α-connections) that preserve the metric under parallel transport by their joint actions; and (iii) a family of divergence functions ( α-divergence) defined on Μθ x Μθ, which induce the metric and the dual connections. Here, we construct an extension of this differential geometric structure from Μθ (that of parametric probability density functions) to the manifold, Μ, of non-parametric functions on X, removing the positivity and normalization constraints. The generalized Fisher information and α-connections on M are induced by an α-parameterized family of divergence functions, reflecting the fundamental convex inequality associated with any smooth and strictly convex function. The infinite-dimensional manifold, M, has zero curvature for all these α-connections; hence, the generally non-zero curvature of M can be interpreted as arising from an embedding of Μθ into Μ. Furthermore, when a parametric model (after a monotonic scaling) forms an affine submanifold, its natural and expectation parameters form biorthogonal coordinates, and such a submanifold is dually flat for α = ± 1, generalizing the results of Amari’s α-embedding. The present analysis illuminates two different types of duality in information geometry, one concerning the referential status of a point (measurable function) expressed in the divergence function (“referential duality”) and the other concerning its representation under an arbitrary monotone scaling (“representational duality”).

Examples of Application of Nonparametric Information Geometry to Statistical Physics

Nonparametric Information Geometry: From Divergence Function to Referential-Representational Biduality on Statistical Manifolds

Information Geometry of the Gaussian Space

Dually affine Information Geometry modeled on a Banach space

Information Geometry for the Working Information Theorist

Monte Carlo Information Geometry: The dually flat case

Progress in Theory and Applications of Information Geometry

Information Geometric Complexity of Entropic Motion on Curved Statistical Manifolds under Different Metrizations of Probability Spaces

Differential Geometry for Model Independent Analysis of Images and Other Non-Euclidean Data: Recent Developments

Information Newton's flow: second-order optimization method in probability space

Information geometric methods for complexity

Geometric Modeling in Probability and Statistics

Information Geometry of Dynamics on Graphs and Hypergraphs

Statistical Physics through the Lens of Real-Space Mutual Information

High-dimensional non-convex landscapes and gradient descent dynamics

Information Geometry and Statistical Manifold

The nonparametric Fisher geometry and the chi-square process density prior

Affine statistical bundle modeled on a Gaussian Orlicz-Sobolev space

On some information-theoretic aspects of non-linear statistical inverse problems

Asymmetric Relaxations Through the Lens of Information Geometry

Understanding Higher-Order Interactions in Information Space