Luigi AcerbiApr 5, 2023

1/ PyVBMC 1.0 is out! 🎉

https://github.com/acerbilab/pyvbmc

A new Python package for efficient Bayesian inference.

Get a posterior distribution over model parameters + the model evidence with a small number of likelihood evaluations.

No AGI was created in the process!

GitHub - acerbilab/pyvbmc: PyVBMC: Variational Bayesian Monte Carlo algorithm for posterior and model inference in Python

PyVBMC: Variational Bayesian Monte Carlo algorithm for posterior and model inference in Python - GitHub - acerbilab/pyvbmc: PyVBMC: Variational Bayesian Monte Carlo algorithm for posterior and mode...

GitHub
Show threadCharles Driver
@AcerbiLuigi looks nice. Does it approximate worse / perform slower / both, as dimensions increase?
Apr 6, 2023 at 8:54amWeb
Show threadLuigi AcerbiApr 6, 2023

@CharlesDriverAU Both! 😀

But that's kind of inevitable for any inference method, especially if you don't require gradients.

In practice, that's why I'd recommend it for inference up to ~10 parameters, more than that is pushing it (although it might still work; my PhD student showed me that for a near-Gaussian posterior it worked fine in 26 dimensions).