EM++: A parameter learning framework for stochastic switching systems

TL;DR

This paper introduces EM++, a novel parameter learning framework for stochastic switching systems, extending EM algorithms with proven global convergence and validated through extensive experiments.

Contribution

It presents a general switching dynamical system model and a majorization-minimization-based algorithm EM++, with theoretical convergence guarantees.

Findings

EM++ reduces to EM for Gaussian distributions

Global convergence of EM++ is proven

Numerical experiments validate effectiveness

Abstract

This paper proposes a general switching dynamical system model, and a custom majorization-minimization-based algorithm EM++ for identifying its parameters. For certain families of distributions, such as Gaussian distributions, this algorithm reduces to the well-known expectation-maximization method. We prove global convergence of the algorithm under suitable assumptions, thus addressing an important open issue in the switching system identification literature. The effectiveness of both the proposed model and algorithm is validated through extensive numerical experiments.