Calibration Loss

Hello fellow Bayesians,

were/are there any attempts towards integrating a calibration loss, e.g., as proposed in [2310.13402] Calibrating Neural Simulation-Based Inference with Differentiable Coverage Probability , into Bayesflow? I am currently experimenting with this and wonder what the community’s take on this approach is.

Greetings, Matze

Hi Matze, the feature is not planned yet but I would be happy to hear about your experience with the method outside the results in the paper.

I will post the repository here once I have a working proof of concept.