Answers are grounded in this paper and cite their sources.
This paper presents a probabilistic perspective on iterative methods for approximating the solution x_* ∈ Rᵈ of a nonsingular linear system A x_* = b. In the approach a standard iterative method on Rᵈ is lifted to act on the space of probability distributions P(Rᵈ). Classically, an iterative method produces a sequence xₘ of approximations that converge to x_*. The output of the iterative methods proposed in this paper is, instead, a sequence of probability distributions μₘ ∈ P(Rᵈ). The distributional output both provides a "best guess" for x_*, for example as the mean of μₘ, and also probabilistic uncertainty quantification for the value of x_* when it has not been exactly determined. Theoretical analysis is provided in the prototypical case of a stationary linear iterative method. In this setting we characterise both the rate of contraction of μₘ to an atomic measure on x_* and the nature of the uncertainty quantification being provided. We conclude with an empirical illustration that highlights the insight into solution uncertainty that can be provided by probabilistic iterative methods.
No takes yet. Share an insight, caveat, or question.
Cockayne et al. (2020) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: