[Ml-stat-talks] matt hoffman, wednesday 12:30pm

David Blei blei at CS.Princeton.EDU
Mon May 14 16:23:21 EDT 2012


(possibly resent.)

alumnus matt hoffman speaks on wednesday at 12:30pm in CS 402.  please
come, especially if you care about approximate inference (or pizza).
title/abstract are below.



The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian
Monte Carlo

Matt Hoffman, Columbia University
CS Room 402

Hamiltonian Monte Carlo (HMC) is a Markov Chain Monte Carlo (MCMC)
algorithm that avoids the random walk behavior and sensitivity to
correlations that plague many MCMC methods by taking a series of steps
informed by first-order gradient information. These features allow it
to converge to high-dimensional target distributions much more quickly
than popular methods such as random walk Metropolis or Gibbs sampling.
However, HMC's performance is highly sensitive to two user-specified
parameters: a step size epsilon and a desired number of steps L. In
particular, if L is too small then the algorithm exhibits undesirable
random walk behavior, while if L is too large the algorithm wastes
computation. We present the No-U-Turn Sampler (NUTS), an extension to
HMC that eliminates the need to set a number of steps L. NUTS uses a
recursive algorithm to build a set of likely candidate points that
spans a wide swath of the target distribution, stopping automatically
when it starts to double back and retrace its steps. NUTS is able to
achieve similar performance to a well tuned standard HMC method,
without requiring user intervention or costly tuning runs. NUTS can
thus be used in applications such as BUGS-style automatic inference
engines that require efficient "turnkey'' sampling algorithms.

More information about the Ml-stat-talks mailing list