Flexible density estimator for prior

[florianhartig]

It would be nice to have a more flexible density estimator for the createPriorDensity function.

Possible options

• gaussian random fields

• https://cran.r-project.org/web/packages/esaddle/

• any other ML algorithm that gives good results

• local density approximation!

Check for how this scales with the number of parameters

[florianhartig]

Create a bunch of test functions with different # parameters and test!

[florianhartig]

Hi Max, have a look at this https://www.biogeosciences-discuss.net/bg-2018-96/ for the solution with Gaussian random fields

[florianhartig]

Alternative would be the https://cran.r-project.org/web/packages/esaddle/index.html package