q-Paths: Generalizing the Geometric Annealing Path using Power Means

This is the code accompanying q-Paths: Generalizing the Geometric Annealing Path using Power Means. Code for the ais experiments is available in qpaths_uai_2021/ais/, and code for smc experiments in qpaths_uai_2021/smc/, with run instructions contained within each folder. Code for all visualizations is available in qpaths_uai_2021/visualizations/.

q-path helper functions to be used across projects are available here.

@inproceedings{masrani2021qpaths,
  title={q-Paths: Generalizing the Geometric Annealing Path using Power Means},
  author={Masrani, Vaden and Brekelmans, Rob and Bui, Thang and Nielsen, Frank  and Galstyan, Aram and Steeg, Greg Ver and Wood, Frank},
  booktitle={Proceedings of the 37th Conference on Uncertainty in Artificial Intelligence},
  year={2021}
}

Acknowledgements

We thank Xuechen Li for his BDMC package used in qpaths_uai_2021/ais/ and Nicolas Chopin for the particles package used in qpaths_uai_2021/smc/.