Probabilistic Numerics for Differential Equations

The field of Probabilistic Numerics aims to quantify numerical uncertainty arising from finite computational resources. By treating the solution of an ordinary differential equation (ODE) as a problem of Bayesian inference, probabilistic numerical ODE solvers return a posterior probability distribution over ODE solutions. This poster presents ProbNumDiffEq.jl, a package for probabilistic numerical solvers for ODEs, built on OrdinaryDiffEq.jl.

The speaker’s profile picture
Nathanael Bosch

I'm a PhD student at University of Tübingen, working on probabilistic numerics, machine learning, and Bayesian inference.