Graduate School and Research Center in Digital Sciences

Variational calibration of computer models

Marmin, Sébastien; Filippone, Maurizio

Submitted on HAL and ArXiv, 29 October 2018

Bayesian calibration of black-box computer models offers an established framework to obtain a posterior distribution over model parameters. Traditional Bayesian calibration involves the emulation of the computer model and an additive model discrepancy term using Gaussian processes; inference is then carried out using MCMC. These choices pose computational and statistical challenges and limitations, which we overcome by proposing the use of approximate Deep Gaussian processes and variational inference techniques. The result is a practical and scalable framework for calibration, which obtains competitive performance compared to the state-of-the-art.

Document Arxiv Hal Bibtex

Title:Variational calibration of computer models
Department:Data Science
Eurecom ref:5722
Copyright: © EURECOM. Personal use of this material is permitted. The definitive version of this paper was published in Submitted on HAL and ArXiv, 29 October 2018 and is available at :
Bibtex: @inproceedings{EURECOM+5722, year = {2018}, title = {{V}ariational calibration of computer models}, author = {{M}armin, {S}{\'e}bastien and {F}ilippone, {M}aurizio}, booktitle = {{S}ubmitted on {HAL} and {A}r{X}iv, 29 {O}ctober 2018}, address = {}, month = {10}, url = {} }
See also: