Learning Bayesian Statistics cover image

#78 Exploring MCMC Sampler Algorithms, with Matt D. Hoffman

Learning Bayesian Statistics

00:00

Generalized HMC

One possible application for the generalized HMC strategy that, you know, it's something it really does make easier to do than. There could be situations where there are computational reasons it's much cheaper to update one set of variables or something. The ways to integrate discrete variables into HMC are they exist, but they're definitely not perfect. But if you take smaller steps and update those discrete variables more frequently, then essentially you're still getting this random noise,. If you take 10 steps with one 10th, the step size, there's sort of a continuous limit where that looks a lot like averaging 10 of these independent gradients.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app