5
$\begingroup$

A couple of years ago I learned about recent work in parallelizing slice sampling methods. More recently, I have read great things about NUTS and Hamiltonian Monte Carlo methods (HMC) in general (e.g. pymc3 uses NUTS) This makes me wonder:

  • What are NUTS and HMC particularly good for? Can they benefit from parallelization as much as, e.g. slice sampling?
  • How do they compare to elliptical slice sampling and other methods in general?
Improve this question
$\endgroup$
1
  • 1
    $\begingroup$ This is not a direct answer to your question, but HMC and NUTS are often used in multilevel models, where you encounter high correlation among parameters at the (many) levels of model hierarchy. $\endgroup$
    – Sycorax ♦
    Aug 14, 2014 at 14:51

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.