Computing Lower and Upper Expectations under Epistemic Independence

This papers investigates the computation of lower/upper expectations that must cohere with a collection of probabilistic assessments and a collection of judgements of epistemic independence. New algorithms, based on multilinear programming, are presented, both for independence among events and among random variables. Separation properties of graphical models are also investigated. Key words: Sets of probability measures, concepts of independence, imprecise probabilities, epistemic independence, multilinear programming

• Type: Conference
• Year: 2005
• Where: ISIPTA
• Authors: Cassio Polpo de Campos, Fabio Gagliardi Cozman
• Keywords: Epistemic Independence, ISIPTA 2005, Lower/upper Expectations, Mathematics, Multilinear Programming