/ExplainingGaussianProcess

Primary LanguageJupyter NotebookMIT LicenseMIT

GP-SHAP: Explaining Gaussian Process Models

figure

A simple illustration to run RKHS-SHAP, GP-SHAP, and BayesGPSHAP to explain kernel methods and Gaussian process.

The algorithms used in this repo came primarily out of the following papers. If you use RKHS-SHAP or GP-SHAP in your research we would appreciate a citation to the appropriate paper(s):

@article{chau2022rkhs,
  title={RKHS-SHAP: Shapley values for kernel methods},
  author={Chau, Siu Lun and Hu, Robert and Gonzalez, Javier and Sejdinovic, Dino},
  journal={Advances in Neural Information Processing Systems},
  volume={35},
  pages={13050--13063},
  year={2022}
}

@article{chau2023explaining,
  title={Explaining the Uncertain: Stochastic Shapley Values for Gaussian Process Models},
  author={Chau, Siu Lun and Muandet, Krikamol and Sejdinovic, Dino},
  journal={arXiv preprint arXiv:2305.15167},
  year={2023}
}