Probabilistic Numerics and Uncertainty in Computations

Abstract

We deliver a call to arms for probabilistic numerical methods: algorithms for numerical tasks, including linear algebra, integration, optimization and solving differential equations, that return uncertainties in their calculations. Such uncertainties, arising from the loss of precision induced by numerical calculation with limited time or hardware, are important for much contemporary science and industry. Within applications such as climate science and astrophysics, the need to make decisions on the basis of computations with large and complex data has led to a renewed focus on the management of numerical uncertainty. We describe how several seminal classic numerical methods can be interpreted naturally as probabilistic inference. We then show that the probabilistic view suggests new algorithms that can flexibly be adapted to suit application specifics, while delivering improved empirical performance. We provide concrete illustrations of the benefits of probabilistic numeric algorithms on real scientific problems from astrometry and astronomical imaging, while highlighting open problems with these new algorithms. Finally, we describe how probabilistic numerical methods provide a coherent framework for identifying the uncertainty in calculations performed with a combination of numerical algorithms (e.g. both numerical optimisers and differential equation solvers), potentially allowing the diagnosis (and control) of error sources in computations.

Author Generated Postprint. 17 pages, 4 Figures, 1 Table

References in corpus (2)

• Astrometric Detection of Giant Planets Around Nearby M Dwarfs: The Gaia Potential
• Gaussian Belief Propagation Solver for Systems of Linear Equations

Cited by in corpus (17)

• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• Inferring solutions of differential equations using noisy multi-fidelity data
• Newton vs the machine: solving the chaotic three-body problem using deep neural networks
• Probabilistic Numerical Methods for PDE-constrained Bayesian Inverse Problems
• A Modern Retrospective on Probabilistic Numerics
• Efficient acquisition rules for model-based approximate Bayesian computation
• Strong convergence rates of probabilistic integrators for ordinary differential equations
• MEMe: An Accurate Maximum Entropy Method for Efficient Approximations in Large-Scale Machine Learning
• Bayesian Numerical Methods for Nonlinear Partial Differential Equations
• From measurements to inferences of physical quantities in numerical simulations
• Randomised one-step time integration methods for deterministic operator differential equations
• Consistency and convergence of simulation schemes in Information field dynamics
• Bayesian Methods for Exoplanet Science
• Radiative Transfer as a Bayesian Linear Regression problem
• An importance sampling approach for reliable and efficient inference in Bayesian ordinary differential equation models
• Quasinormal modes from numerical relativity with Bayesian inference
• Modelling pathwise uncertainty of Stochastic Differential Equations samplers via Probabilistic Numerics