Quantifier Elimination for Statistical Problems

Abstract

Recent improvement on Tarski's procedure for quantifier elimination in the first order theory of real numbers makes it feasible to solve small instances of the following problems completely automatically: 1. listing all equality and inequality constraints implied by a graphical model with hidden variables. 2. Comparing graphyical models with hidden variables (i.e., model equivalence, inclusion, and overlap). 3. Answering questions about the identification of a model or portion of a model, and about bounds on quantities derived from a model. 4. Determing whether a given set of independence assertions. We discuss the foundation of quantifier elimination and demonstrate its application to these problems.

Appears in Proceedings of the Fifteenth Conference on Uncertainty in Artificial Intelligence (UAI1999)

References in corpus (2)

• Strong Completeness and Faithfulness in Bayesian Networks
• Graphical Models and Exponential Families

Cited by in corpus (14)

• Machine learning non-local correlations
• Semidefinite tests for quantum network topologies
• Experimental Test of Nonlocal Causality
• Causal inference via algebraic geometry: feasibility tests for functional causal structures with two binary observed variables
• The Inflation Technique Completely Solves the Causal Compatibility Problem
• Inferring latent structures via information inequalities
• Semidefinite tests for latent causal structures
• A convergent inflation hierarchy for quantum causal structures
• Polynomial Constraints in Causal Bayesian Networks
• Undecidability of Network Coding, Conditional Information Inequalities, and Conditional Independence Implication
• Monogamy of Temporal Correlations: Witnessing non-Markovianity Beyond Data Processing
• Quantum Causal Structure and Quantum Thermodynamics
• Entropic Inequality Constraints from $e$-separation Relations in Directed Acyclic Graphs with Hidden Variables
• Partial Identifiability in Discrete Data With Measurement Error