Evidence estimation using stochastic likelihood approximations

Cameron, Scott (2020-04)

Thesis (MSc)--Stellenbosch University, 2020.

Thesis

ENGLISH ABSTRACT: We consider the problem of estimating evidence for parametric Bayesian models in the large data regime. Many existing evidence estimation algorithms scale poorly due to their need to repeatedly calculate the exact likelihood, which requires iterating over the entire data set. This inefficiency can be circumvented with the use of stochastic likelihood estimates on small sub-samples of the data set. We therefore tackle this problem by introducing stochastic gradient Monte Carlo methods for evidence estimation, our main contribution being stochastic gradient annealed importance sampling. Our approach enables efficient online evidence estimation for large data sets. SGAIS is considerably faster than previous approaches for single data sets, with improved order complexity for online estimation, without noticeable loss of accuracy.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/108005
This item appears in the following collections: