Jon Cockayne

Associate Professor at University of Southampton

Publications

Working

Working

Constructive Disintegration and Conditional Modes
Nathaël Da Costa, Marvin Pförtner, Jon Cockayne
arXiv

2025

2025

Learning to Solve Related Linear Systems
Disha Hegde, Jon Cockayne
Proceedings of the First International Conference on Probabilistic Numerics
arXiv journal
Randomised Postiterations for Calibrated BayesCG
Niall Vyas, Disha Hegde, Jon Cockayne
Proceedings of the First International Conference on Probabilistic Numerics
arXiv journal
Calibrated Computation-Aware Gaussian Processes
Disha Hegde, Mohamed Adil, Jon Cockayne
AISTATS 2025
arXiv journal
Computation-Aware Kalman Filtering and Smoothing
Marvin Pförtner, Jonathan Wenger, Jon Cockayne, Philipp Hennig
AISTATS 2025
arXiv journal

2023

2023

Theoretical Guarantees for the Statistical Finite Element Method
Yanni Papandreou, Jon Cockayne, Mark Girolami, Andrew Duncan
SIAM Journal of Uncertainty Quantification
arXiv
A Probabilistic Taylor Expansion with Applications in Filtering and Differential Equations
Toni Karvonen, Jon Cockayne, Filip Tronarp, Simo Särkkä
Transactions of Machine Learning Research
arXiv
Statistical Properties of BayesCG Under the Krylov Prior
Tim Reid, Ilse Ipsen, Jon Cockayne, Chris Oates
Numerische Mathematik
arXiv

2022

2022

Testing whether a Learning Procedure is Calibrated
Jon Cockayne, Matthew Graham, Chris Oates, Tim Sullivan
Journal of Machine Learning Research
arXiv journal
Radiative Transfer as a Bayesian Linear Regression Problem
Frederik De Ceuster, Thomas Ceulemans, Jon Cockayne, Leen Decin, Jeremy Yates
Monthly Notices of the Royal Astronomical Society
arXiv

2021

2021

Probabilistic Gradients for Fast Calibration of Differential Equation Models
Jon Cockayne, Andrew Duncan
SIAM Journal of Uncertainty Quantification
arXiv
Probabilistic Iterative Methods for Linear Systems
Jon Cockayne, Ilse Ipsen, Chris Oates, Tim Reid
Journal of Machine Learning Research
arXiv journal
Bayesian Numerical Methods for Nonlinear Partial Differential Equations
Junyang Wang, Jon Cockayne, Oksana Chkrebtii, Tim Sullivan, Chris Oates
Statistics and Computing
arXiv journal
Optimal Thinning of MCMC Output
Marina Riabiz, Wilson Chen, Jon Cockayne, Pawel Swietach, Steven Niederer, Lester Mackey, Chris Oates
Journal of the Royal Statistical Society Series B
arXiv

2020

2020

BayesCG As An Uncertainty Aware Version of CG
Tim Reid, Ilse Ipsen, Jon Cockayne, Chris Oates
arXiv
Optimality Criteria for Probabilistic Numerical Methods
Chris Oates, Jon Cockayne, Dennis Prangle, Tim Sullivan, Mark Girolami
Multivariate Algorithms and Information-Based Complexity
arXiv journal
A Role for Symmetry in the Bayesian Solution of Differential Equations
Junyang Wang, Jon Cockayne, Chris Oates
Bayesian Analysis
arXiv journal

2019

2019

A Bayesian Conjugate Gradient Method
Jon Cockayne, Chris Oates, Ilse Ipsen, Mark Girolami
Bayesian Analysis, with discussion and rejoinder
arXiv journal BA Webinar
Bayesian Probabilistic Numerical Methods
Jon Cockayne, Chris Oates, Tim Sullivan, Mark Girolami
SIAM Review
arXiv journal
Bayesian Probabilistic Numerical Methods in Time-Dependent State Estimation for Industrial Hydrocyclone Equipment
Chris Oates, Jon Cockayne, Robert Aykroyd, Mark Girolami
Journal of the American Statistical Association
arXiv journal
Probabilistic Linear Solvers: A Unifying View
Simon Bartels, Jon Cockayne, Ilse Ipsen, Philipp Hennig
Statistics and Computing
arXiv journal
Convergence Rates for a Class of Estimators Based on Stein's Method
Chris Oates, Jon Cockayne, Francois-Xavier Briol, Mark Girolami
Bernoulli
arXiv journal

2017

2017

On the Sampling Problem for Kernel Quadrature
Francois-Xavier Briol, Chris Oates, Jon Cockayne, Wilson Chen, Mark Girolami
Proceedings of the 34th International Conference on Machine Learning
arXiv journal
On the Bayesian Solution of Differential Equations
Junyang Wang, Jon Cockayne, Chris Oates
Proceedings of the 38th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering
arXiv

2016

2016

Probabilistic Numerical Methods for PDE-Constrained Bayesian Inverse Problems
Jon Cockayne, Chris Oates, Tim Sullivan, Mark Girolami
Proceedings of the 38th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering
arXiv journal

Thesis

My PhD was awarded by the University of Warwick in 2019. The title of my thesis was Probabilistic Numerical Methods; it can be found here.

Software

Please visit my github to find all of the software that I have made available from my research.

Libraries

bayesian_pdes
Implementation of the PDE solver from "Probabilistic Meshless Methods".
BCG
An implementation of the BayesCG algorithm from "A Bayesian Conjugate Gradient Method".
mcmc
Implementations of various MCMC routines in pure Python.
cmcmc
Implementations of various MCMC routines in C++.

Code from Papers

hydrocyclone_code
Code for reproducing results from "Bayesian PNM for Industrial Hydrocyclone Equipment".

Contact

  [email protected]