A fast sparse Bayesian learning algorithm based on gaussian scale mixture model for regression problem
This code is for paper titled "An efficient sparse Bayesian learning algorithm Based on Gaussian-scale mixtures".
The images in dataset are acquired from http://sparselab.stanford.edu/ and http://decsai.ugr.es/cvg/dbimagenes/.
The function FastLaplace.m in tools corresponds to the fast SBL algorithm based on Laplace priors, which is acquired from the origianl authors at http://www.dbabacan.info/publications.html. The paper is titled "Bayesian Compressive Sensing using Laplace Priors".
GGAMP-SBL.m corresponds to the algorithm 1 in paper titled "A GAMP based low complexity sparse bayesian learning algorithm".
For comparison, sparseLab 2.1 and RVM V1.1 toolboxs are needed, which can be obtained from http://sparselab.stanford.edu/ and http://www.miketipping.com/downloads.htm, respectively.
This code is implemented in Matlab 2019b. If have any questions, please contact [email protected]
If you use any part of our codes, please cite our paper.
W. Zhou, H. -T. Zhang and J. Wang, "An Efficient Sparse Bayesian Learning Algorithm Based on Gaussian-Scale Mixtures," IEEE Transactions on Neural Networks and Learning Systems, doi: 10.1109/TNNLS.2020.3049056. https://ieeexplore.ieee.org/abstract/document/9334415
bibtex:
@ARTICLE{zhou2021efficient, author={W. {Zhou} and H.-T. {Zhang} and J. {Wang}}, journal={IEEE Transactions on Neural Networks and Learning Systems}, title={An Efficient Sparse Bayesian Learning Algorithm Based on Gaussian-Scale Mixtures}, year={2021}, volume={}, number={}, pages={1-14}, doi={10.1109/TNNLS.2020.3049056}}