Information trajectory of optimal learning

The paper outlines some basic principles of a geometric and
non-asymptotic theory of learning systems. An evolution of such a system is represented by points on a statistical manifold, and a topology related to information dynamics is introduced to define trajectories continuous in information. It is shown that optimization of learning with respect to a given utility function leads to an evolution described by a continuous trajectory. Path integrals along the trajectory define the optimal utility and information bounds. Closed form expressions are derived for two important types of utility functions. The presented approach is a generalization of the use of Orlicz spaces in information geometry, and it gives a new, geometric interpretation of the classical information value theory and statistical mechanics. In addition, theoretical predictions are evaluated experimentally by comparing performance of agents learning in a non-stationary stochastic environment.