Irrelevance-based partial MAPs are useful constructs for domain-independent explanation using belief networks. We look at two definitions for such partial MAPs, and prove important properties that are useful in designing algorithms for computing them effectively. We make use of these properties in modifying our standard MAP best-first algorithm, so as to handle irrelevance-based partial MAPs.
|Original language||English GB|
|Title of host publication||Uncertainty Proceedings 1991 Proceedings of the Seventh Conference (1991)|
|State||Published - 1991|
|Event||Seventh Conference on Uncertainty in Artificial Intelligence - University of California at Los Angeles (UCLA), Los Angeles, United States|
Duration: 13 Jul 1991 → 15 Jul 1991
|Conference||Seventh Conference on Uncertainty in Artificial Intelligence|
|Period||13/07/91 → 15/07/91|