Fast Dual Subgradient Optimization of the Integrated Transportation Distance Between Stochastic Kernels

TL;DR

This paper introduces a novel, efficient dual subgradient algorithm for approximating the integrated transportation distance between stochastic kernels, enabling practical analysis of Markov systems with limited computational resources.

Contribution

It presents a new approximation method for the integrated transportation distance and a specialized dual algorithm that avoids costly matrix operations for practical implementation.

Findings

Effective approximation of stochastic kernels demonstrated

Algorithm achieves fast and efficient kernel construction

Practical utility shown through illustrative examples

Abstract

A generalization of the Wasserstein metric, the integrated transportation distance, establishes a novel distance between probability kernels of Markov systems. This metric serves as the foundation for an efficient approximation technique, enabling the replacement of the original system's kernel with a kernel with a discrete support of limited cardinality. To facilitate practical implementation, we present a specialized dual algorithm capable of constructing these approximate kernels quickly and efficiently, without requiring computationally expensive matrix operations. Finally, we demonstrate the efficacy of our method through several illustrative examples, showcasing its utility in practical scenarios. This advancement offers new possibilities for the streamlined analysis and manipulation of stochastic systems represented by kernels.