Roadmap

The goal of this repository is to collect high-dimensional volume estimation algorithms in a way that is hot swappable for benchmarks. i.e given a membership function, in theory we should be able to get an unbiased estimate for volumes.

The current version uses pigeons.jl with parallel tempering in the spirit of basin volume calculations in computational chemistry/physics

https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.106.245502

Note that there are developments elswhere, see volesti

https://github.com/GeomScale/volesti

Also see this thesis by ben cousins and vempala's work

https://aco.gatech.edu/sites/default/files/documents/cousins_ben_thesis.pdf

https://dl.acm.org/doi/epdf/10.1145/3795687