Assume a reduction of complexity as dimensionality increases through prior

[odunbar]

Issue

For our current ad-hoc priors for the RF calibration parameters

CalibrateEmulateSample.jl/src/RandomFeature.jl

Line 246 in eb262cb

function build_default_prior(name::SS, n_hp::Int, df::DiagonalFactor) where {SS <: AbstractString}

It may make sense to draw some inspiration from work such as this where one takes less rough priors as (effective) dimensionality increases. This may make parameter spaces more robust to changing problem dimension