CUED Publications database

Gauged Mini-Bucket Elimination for Approximate Inference

Ahn, S and Chertkov, M and Shin, J and Weller, A Gauged Mini-Bucket Elimination for Approximate Inference. (Unpublished)

Full text not available from this repository.


Computing the partition function $Z$ of a discrete graphical model is a fundamental inference challenge. Since this is computationally intractable, variational approximations are often used in practice. Recently, so-called gauge transformations were used to improve variational lower bounds on $Z$. In this paper, we propose a new gauge-variational approach, termed WMBE-G, which combines gauge transformations with the weighted mini-bucket elimination (WMBE) method. WMBE-G can provide both upper and lower bounds on $Z$, and is easier to optimize than the prior gauge-variational algorithm. We show that WMBE-G strictly improves the earlier WMBE approximation for symmetric models including Ising models with no magnetic field. Our experimental results demonstrate the effectiveness of WMBE-G even for generic, nonsymmetric models.

Item Type: Article
Uncontrolled Keywords: stat.ML stat.ML
Divisions: Div F > Computational and Biological Learning
Depositing User: Cron Job
Date Deposited: 24 Jan 2018 20:07
Last Modified: 18 Feb 2021 18:14