The other is Cari Kaufman, who builds probability distributions over the results of a climate simulation. (the idea seems to be to extrapolate from simulations actually run with similar but not identical parameters)
I was introduced to the idea of ‘emulation’ of complex models by Tony O’Hagan a few years back, where you use a Gaussian Process to model what a black box simulation will give across all possible inputs, seeded with actual simulation runs that you performed. (This also helps with active learning, in that you can find the regions of the input space where you’re most uncertain what the simulation will give, and then run a simulation with those input parameters.) I believe the first application it saw was also in climate modeling.
Do you know of any cases where this simulation-seeded Gaussian Process was then used as a prior, and updated on empirical data?
Like...
uncertain parameters—simulation--> distribution over state
noisy observations—standard bayesian update--> refined distribution over state
Cari Kaufman’s research profile made me think that’s something she was interested in. But I haven’t found any publications by her or anyone else that actually do this.
I actually think that I misread her research description, latching on to the one familiar idea.
I was introduced to the idea of ‘emulation’ of complex models by Tony O’Hagan a few years back, where you use a Gaussian Process to model what a black box simulation will give across all possible inputs, seeded with actual simulation runs that you performed. (This also helps with active learning, in that you can find the regions of the input space where you’re most uncertain what the simulation will give, and then run a simulation with those input parameters.) I believe the first application it saw was also in climate modeling.
Do you know of any cases where this simulation-seeded Gaussian Process was then used as a prior, and updated on empirical data?
Like...
uncertain parameters—simulation--> distribution over state
noisy observations—standard bayesian update--> refined distribution over state
Cari Kaufman’s research profile made me think that’s something she was interested in. But I haven’t found any publications by her or anyone else that actually do this.
I actually think that I misread her research description, latching on to the one familiar idea.
None come to mind, sadly. :( (I haven’t read through all of his work, though, and he might know someone who took it in that direction.)