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 gambles. Separation properties of graphical models are also investigated.
Keywords. Lower and upper expectations, credal sets, credal networks, multilinear programming
The paper is availabe in the following formats:
São Paulo, SP, Brazil
Av. Prof. Mello Moraes, 2231
Cidade Univesitaria, CEP 05508-900
Sao Paulo, SP - BRAZIL