Probabilistic Model Validation for Uncertain Nonlinear Systems

A. Halder and R. Bhattacharya

Automatica, accepted, 2014.

Abstract: This paper presents a probabilistic model validation methodology for nonlinear systems. The proposed formulation is simple, intuitive, and accounts both deterministic and stochastic nonlinear systems with parametric and nonparametric uncertainties. Instead of hard invalidation methods available in the literature, a relaxed notion of validation in probability is introduced. To guarantee provably correct inference, algorithm for constructing probabilistically robust validation certificate is given along with computational complexities. Some examples are worked out to illustrate its use.

  • Source codes
         – Code for comparing the accuracy of empirical Wasserstein computation (finite-dimensional LP) against the exact Wasserstein computation (infinite dimensional LP):
         – Function to compute Wassertein distance between multivariate normal distributions:
         – TBD
         – TBD