Belief revision with unreliable observations

Craig Boutilier*, Nir Friedman, Joseph Y. Halpern

*Corresponding author for this work

Research output: Contribution to conferencePaperpeer-review

28 Scopus citations


Research in belief revision has been dominated by work that lies firmly within the classic AGM paradigm, characterized by a well-known set of postulates governing the behavior of `rational' revision functions. A postulate that is rarely criticized is the success postulate: the result of revising by an observed proposition φ results in belief in φ. This postulate, however, is often undesirable in settings where an agent's observations may be imprecise or noisy. We propose a semantics that captures a new ontology for studying revision functions, which can handle noisy observations in a natural way while retaining the classical AGM model as a special case. We present a characterization theorem for our semantics, and describe a number of natural special cases that allow ease of specification and reasoning with revision functions. In particular, by making the Markov assumption, we can easily specify and reason about revision.

Original languageAmerican English
Number of pages8
StatePublished - 1998
Externally publishedYes
EventProceedings of the 1998 15th National Conference on Artificial Intelligence, AAAI - Madison, WI, USA
Duration: 26 Jul 199830 Jul 1998


ConferenceProceedings of the 1998 15th National Conference on Artificial Intelligence, AAAI
CityMadison, WI, USA


Dive into the research topics of 'Belief revision with unreliable observations'. Together they form a unique fingerprint.

Cite this