Tanimoto Random Features for Scalable Molecular Machine Learning

paper
Author

Austin Tripp, Sergio Bacallado, Sukriti Singh, José Miguel Hernández-Lobato

Official version Version I consider authoritative

Summary

Scalable approximation for the Tanimoto kernel (useful for GPs on molecules).

My contribution

Initiated project, co-developed minmax random features, did writing + experiments.

Thoughts

(as of 2023-11-24)

This paper does what it says it does: provides an approximation to this kernel. I am happy that this paper made modest claims and demonstrated them, in contrast to many papers which support bold claims with shaky experiments.