Incremental Probabilistic Inference

TL;DR

This paper introduces an incremental probabilistic inference system that operates at a finer granularity, enabling more efficient and adaptable probabilistic reasoning akin to propositional truth maintenance systems.

Contribution

It proposes a novel approach to probabilistic inference by focusing on smaller inference tasks, enhancing incrementality and serving as a low-level probabilistic representation service.

Findings

Enables incremental probabilistic inference at a finer granularity.

Supports interleaved problem-model construction and evaluation.

Improves flexibility and efficiency of probabilistic reasoning.

Abstract

Propositional representation services such as truth maintenance systems offer powerful support for incremental, interleaved, problem-model construction and evaluation. Probabilistic inference systems, in contrast, have lagged behind in supporting this incrementality typically demanded by problem solvers. The problem, we argue, is that the basic task of probabilistic inference is typically formulated at too large a grain-size. We show how a system built around a smaller grain-size inference task can have the desired incrementality and serve as the basis for a low-level (propositional) probabilistic representation service.