On the Conditional Independence Implication Problem: A Lattice-Theoretic Approach

08/09/2014
by   Mathias Niepert, et al.
0

A lattice-theoretic framework is introduced that permits the study of the conditional independence (CI) implication problem relative to the class of discrete probability measures. Semi-lattices are associated with CI statements and a finite, sound and complete inference system relative to semi-lattice inclusions is presented. This system is shown to be (1) sound and complete for saturated CI statements, (2) complete for general CI statements, and (3) sound and complete for stable CI statements. These results yield a criterion that can be used to falsify instances of the implication problem and several heuristics are derived that approximate this "lattice-exclusion" criterion in polynomial time. Finally, we provide experimental results that relate our work to results obtained from other existing inference algorithms.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/20/2013

A Graph-Based Inference Method for Conditional Independence

The graphoid axioms for conditional independence, originally described b...
research
05/09/2012

Logical Inference Algorithms and Matrix Representations for Probabilistic Conditional Independence

Logical inference algorithms for conditional independence (CI) statement...
research
05/30/2021

Lattice Conditional Independence Models and Hibi Ideals

Lattice Conditional Independence models are a class of models developed ...
research
03/27/2023

Inference Rules for Binary Predicates in a Multigranular Framework

In a multigranular framework, the two most important binary predicates a...
research
02/27/2013

Semigraphoids Are Two-Antecedental Approximations of Stochastic Conditional Independence Models

The semigraphoid closure of every couple of CI-statements (GI=conditiona...
research
11/17/2017

Lattices and Their Consistent Quantification

This paper introduces the order-theoretic concept of lattices along with...
research
08/28/2010

Foundations of Inference

We present a simple and clear foundation for finite inference that unite...

Please sign up or login with your details

Forgot password? Click here to reset