There is some progress in that direction though. The bigger problem, as mentioned in the link, it is that that estimator seems to completely break down if you try and use an approximation to the posterior although there seems to be ongoing work to estimate generalisation error just from MCMC samples.
There is some progress in that direction though. The bigger problem, as mentioned in the link, it is that that estimator seems to completely break down if you try and use an approximation to the posterior although there seems to be ongoing work to estimate generalisation error just from MCMC samples.