Maximum a Posteriori Estimation by Search in Probabilistic Programs

Authors

  • David Tolpin University of Oxford
  • Frank Wood Univeristy of Oxford

DOI:

https://doi.org/10.1609/socs.v6i1.18369

Keywords:

probabilistic programming, MAP, MCTS

Abstract

We introduce an approximate search algorithm for fast maximum a posteriori probability estimation in probabilistic programs, which we call Bayesian ascent Monte Carlo (BaMC). Probabilistic programs represent probabilistic models with varying number of mutually dependent finite, countable, and continuous random variables. BaMC is an anytime MAP search algorithm applicable to any combination of random variables and dependencies. We compare BaMC to other MAP estimation algorithms and show that BaMC is faster and more robust on a range of probabilistic models.

Downloads

Published

2021-09-01