Abstract
This paper 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.
Original language | English |
---|---|
Pages (from-to) | 244-260 |
Number of pages | 17 |
Journal | International Journal of Approximate Reasoning |
Volume | 44 |
Issue number | 3 |
DOIs | |
Publication status | Published - 1 Mar 2007 |
Externally published | Yes |
Bibliographical note
Reasoning with Imprecise ProbabilitiesFunding
We thank Peter Walley for sharing with us the algorithm in Fig. A.1. This work has received generous support from HP Brazil R&D. The work has also been supported by CNPq (through grant 3000183/98-4) and FAPESP (through grant 04/09568-0).
Keywords
- Concepts of independence
- Epistemic independence
- Imprecise probabilities
- Multilinear programming
- Sets of probability measures