Betting with probability intervals: the intersection probability

Fabio Cuzzolin
STATUS: submitted to Artificial Intelligence, February 2007; first revision June 2009, to revise Fall 2012

Probability intervals are an attractive tool for reasoning under uncertainty. Unlike belief functions, though, they lack a natural probability transformation (such as the pignistic function for BFs) that could be used for decision making in a utility theory framework. We proposed the use of the intersection probability, a transformation derived originally for belief functions in our geometric framework, as the most natural such transformation. This outlines a novel decision making framework for probability intervals, analogous to the Transferable Belief Model for belief functions.

BibTeX entry

  AUTHOR = "Fabio Cuzzolin", 
  TITLE = "Betting with probability intervals: the intersection probability",
  JOURNAL = "submitted to Artificial Intelligence", 
  YEAR = "2013"