The Fundamental Incompatibility of Hamiltonian Monte Carlo and Data Subsampling — mentioned by tMcGrath

Fri, 10 Nov 2017 09:55:22 GMT

I'd like to flag up this paper on HMC and minibatch inference: https://arxiv.org/abs/1502.01510v1. It seems that SGHMC might not always behave as well as you might like. I should say this isn't to pour cold water on the idea of Bayesian deep learning - I think it's an extremely sensible idea and undoubtedly a good direction for research, but just wanted to caution that this particular MCMC method might have subtle flaws.

Paper and discussion history. A mention is not an endorsement.

The Fundamental Incompatibility of Hamiltonian Monte Carlo and Data Subsampling · HackerPapers discussions

The Fundamental Incompatibility of Hamiltonian Monte Carlo and Data Subsampling — mentioned by tMcGrath