English

Deep Multi-fidelity Gaussian Processes

Machine Learning 2016-04-27 v1 Machine Learning

Abstract

We develop a novel multi-fidelity framework that goes far beyond the classical AR(1) Co-kriging scheme of Kennedy and O'Hagan (2000). Our method can handle general discontinuous cross-correlations among systems with different levels of fidelity. A combination of multi-fidelity Gaussian Processes (AR(1) Co-kriging) and deep neural networks enables us to construct a method that is immune to discontinuities. We demonstrate the effectiveness of the new technology using standard benchmark problems designed to resemble the outputs of complicated high- and low-fidelity codes.

Cite

@article{arxiv.1604.07484,
  title  = {Deep Multi-fidelity Gaussian Processes},
  author = {Maziar Raissi and George Karniadakis},
  journal= {arXiv preprint arXiv:1604.07484},
  year   = {2016}
}
R2 v1 2026-06-22T13:40:43.219Z