Hierarchical Bayesian Regression (HBR) and Analytic Gradient Calculation (GCGP)

Andrej Aderhold, Dirk Husmeier, Marco GrzegorczykUniversity 2016 · DOI

Abstract

The software contains two components: The HBR code and the gradient calculation code (GCGP) that uses the Gaussian Process package GPstuff (http://research.cs.aalto.fi/pml/software/gpstuff/) by: Jarno Vanhatalo, Jaakko Riihimäki, Jouni Hartikainen, Pasi Jylänki, Ville Tolvanen, and Aki Vehtari (2013). GPstuff: Bayesian Modeling with Gaussian Processes. Journal of Machine Learning Research, 14(Apr):1175-1179 The HBR code was originally written by Marco Grzegorczyk, and adapted by Andrej Aderhold for the Biopepa data. It is located in the subfolder 'HBR/', see Section A. for details. The GCGP code was written by Andrej Aderhold, and uses sub-routines from the GPstuff, version 4.4 software that is under the GNU GPL license. It is located in the subfolder GCGP, see Section B. for details.
BibTeX
@inproceedings{aderhold2016hierarchical,
  author = {Andrej Aderhold and Dirk Husmeier and Marco Grzegorczyk},
  title = {{Hierarchical Bayesian Regression (HBR) and Analytic Gradient Calculation (GCGP)}},
  booktitle = {University},
  year = {2016},
  doi = {10.5525/gla.researchdata.351},
}

Loading related content...

Figure