Extending Monte Carlo Methods to Factor Graphs with Negative and Complex Factors

TL;DR

This paper extends Monte Carlo methods to estimate the partition function of factor graphs that include negative and complex factors, broadening their applicability in probabilistic modeling.

Contribution

It introduces novel Monte Carlo techniques capable of handling negative and complex factors in factor graphs, which was not addressed in prior work.

Findings

Successfully estimated partition functions with complex factors

Demonstrated effectiveness on example graphs

Extended Monte Carlo applicability to new classes of problems

Abstract

The partition function of a factor graph can sometimes be accurately estimated by Monte Carlo methods. In this paper, such methods are extended to factor graphs with negative and complex factors.