Experiments with Distributed Anytime Inferencing: Working with Cooperative Algorithms
Edward J. Williams, Eugene Santos, Wright-Patterson Afb, Solomon Eyal Shimony · 1997
Anytime algorithms have demonstrated their useful-ness in solving many classes of intractable and NP-hard problems. This approach allows the potential for improvement in the quality of the solution to be balanced against the cost of generating that improve-ment, both in time and system resources. While sig-nificant work has been accomplished on characterizing individual algorithms and sequences of algorithms, the same has not been done for collections of algorithms. Our research extends the anytime concept by provid-ing feedback to algorithms operating concurrently in a distributed cooperative environment. Traditional any-time algorithms execute individually, taking all their input from the problem being solved; the monitoring task plays little part in the problem solving process.