PSI: Constructing ad-hoc simplices to interpolate high-dimensional unstructured data

Interpolating unstructured data using barycentric coordinates becomes infeasible at high dimensions due to the prohibitive memory requirements of building a Delaunay triangulation. We present new algorithm construct ad-hoc simplices that are empirically guaranteed contain target coordinates, based on nearest neighbor heuristic and an iterative dimensionality reduction through projection. use these interpolate astrophysical cooling function $\Lambda$ show this approach produces good results with just fraction previously required memory.