Technical Working Group

[Bob_Carpenter]

Here’s a relevant recent discussion around VI and diagnostics:

Adjusting default tol_rel_obj threshold in vb

The recent paper by Yao et al. on Stan’s variational inference is pretty clear that lowering the threshold for change in ELBO will improve the fit of the approximated posterior and is more likely to pass the diagnostic checks they propose (see figure 2 in the paper). However, the current default in rstan, option tol_rel_obj, is still set at 0.01. Given the fact that rstan settings for full Bayesian inference are conservative (2000 iterations), it would seem a good practice to change the default …

It’s about adding K-hat diagnostics, changing thresholds on VI runs, etc. I’m not sure it requires a committee meeting to sort out.