Example: pollution

This test shows a surrogate-assisted Bayesian calibration of a time dependent pollution function. Here, the noise will be jointly inferred with the input parameters.

ENVIRONMENTAL MODEL FUNCTION

Authors: Sonja Surjanovic, Simon Fraser University

Derek Bingham, Simon Fraser University

Questions/Comments: Please email Derek Bingham at dbingham@stat.sfu.ca.

Copyright 2013. Derek Bingham, Simon Fraser University.

THERE IS NO WARRANTY, EXPRESS OR IMPLIED. WE DO NOT ASSUME ANY LIABILITY FOR THE USE OF THIS SOFTWARE. If software is modified to produce derivative works, such modified software should be clearly marked. Additionally, this program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; version 2.0 of the License. Accordingly, this program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

For function details and reference information, see:

http://www.sfu.ca/~ssurjano/

This example trains a surrogate and performs Bayesian Inference based on the given data. Active Learning can be activated and will then be performed with Variance Optimal Design based on Entropy and an epsilon-decreasing tradeoff scheme.

Note

This example contains two test_* files that can be run.

Model and Data

Pylink model

Property	Setting
Model type	Function
Number of input parameters	4
Number of output parameters	1
Time- or space- dependency	Yes, ??
MC reference	No

Priors

Parameter	Distribution
M	uniform
D	uniform
L	uniform
tau	uniform

Surrogate

MetaModel settings

Property	Setting
surrogate-type	aPCE
associated model	‘pollution’
degree choices	max degree 8, q-norm truncation 1.0
regression	BCS (Bayesian Compressive Sensing) with 'fast' bootstrap

Training choices

Property	Setting
Static sampling method	latin-hypercube
Number of static samples	150
Number of total samples	150