Posterior error bounds for prior-driven balancing in linear Gaussian inverse problems
Provides theoretical guarantees for model reduction in Bayesian inverse problems, benefiting practitioners who need to balance computational cost and accuracy.
The paper derives a priori error bounds for posterior mean and covariance in linear Gaussian inverse problems when using approximate forward models, specifically applying balanced truncation model order reduction to smoothing problems. The bounds are expressed in terms of truncated Hankel singular values.
In large-scale Bayesian inverse problems, it is often necessary to apply approximate forward models to reduce the cost of forward model evaluations, while controlling approximation quality. In the context of Bayesian inverse problems with linear forward models, Gaussian priors, and Gaussian noise, we use perturbation theory for inverses to bound the error in the approximate posterior mean and posterior covariance resulting from a linear approximate forward model. We then focus on the smoothing problem of inferring the initial condition of linear time-invariant dynamical systems, using finitely many partial state observations. For such problems, and for a specific model order reduction method based on balanced truncation, we show that the impulse response of a certain prior-driven system is closely related to the prior-preconditioned Hessian of the inverse problem. This reveals a novel connection between systems theory and inverse problems. We exploit this connection to prove the first a priori error bounds for system-theoretic model order reduction methods applied to smoothing problems. The bounds control the approximation error of the posterior mean and covariance in terms of the truncated Hankel singular values of the underlying system.