This repo contains the code which was used to compute the results of the paper "Calibrated Adaptive Probabilistic ODE Solvers", presented at AISTATS 2021 (link).
To solve differential equations in Julia with probabilistic numerical solvers, please use
ProbNumDiffEq.jl!
The code in this repository is not meant to be used as generic ODE solvers, whereas ProbNumDiffEq.jl is a Julia package under active development.
It is more stable and documented, its solvers are more efficent, and it contains more features.
The DE solvers it provides are compatible with the
DifferentialEquations.jl
ecosystem.
A Python implementation of these solvers, as well as of additional probabilistic numerical methods, is maintained in ProbNum.
The experiments can be found in ./experiments, and the actual solvers are implemented in ./src.
To start, open a Julia console with julia --project=..
You can then run the experiments with, e.g., include(experiments/1 Stiff Van der Pol/main.jl).
@InProceedings{pmlr-v130-bosch21a,
title = { Calibrated Adaptive Probabilistic ODE Solvers },
author = {Bosch, Nathanael and Hennig, Philipp and Tronarp, Filip},
booktitle = {Proceedings of The 24th International Conference on Artificial Intelligence and Statistics},
pages = {3466--3474},
year = {2021},
editor = {Banerjee, Arindam and Fukumizu, Kenji},
volume = {130},
series = {Proceedings of Machine Learning Research},
month = {13--15 Apr},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v130/bosch21a/bosch21a.pdf},
url = {http://proceedings.mlr.press/v130/bosch21a.html},
}