Kennedy-O'Hagan
Multi-fidelity Gaussian processes: predicting a costly code from cheap approximations
Paper title Predicting the output from a complex computer code when fast approximations are available
Paper doi.org
Combines expensive runs of the most complex version of a computer code with relatively cheap runs of one or more simpler approximations, in a Bayesian approach where prior beliefs about the codes are Gaussian processes, for prediction and uncertainty analysis of the expensive output.
- Published
- Biometrika
Share on LinkedIn, X or Bluesky, or or .
Know related work that belongs here? Suggest a link
Something wrong on this page? Tell the editor