Partial Observability in Computation

Project Description:

Processing information that is partially observable is an important aspect of many AI applications. We have developed a special kind of partially observable Markov decision process (POMDP), called the Single Action POMDP (SA-POMDP). It is our goal to apply SA-POMDPs to various aspects of AI and other forms of computing.

Project Members:

Brett Moore (moore@cs.ttu.edu)
Todd Quasny (quasny@cs.ttu.edu)

Project Papers:

Moore, Quasny, Pyeatt, and Sinzinger. Performance of a Single Action Partially Oberservable Markov Decision Process in a Recognition Task.(ps|pdf) Accepted by ASC 2001 and is scheduled to appear in the proceedings.

Todd Quasny

Last modified: Mon Mar 19 15:36:38 CST 2001