Edward Michael Williams

Learn More
Belief updating in Bayes nets, a well known computationally hard problem, has recently been approximated by several deterministic algorithms, and by various randomized approximation algorithms. Deterministic algorithms usually provide probability bounds, but have an exponential runtime. Some randomized schemes have a polynomial runtime, but provide only(More)
Belief updating in Bayes nets, a well known computationally hard problem, has recently been approximated by several deterministic algorithms, and by various randomized approximation algorithlns. Deterministic algorithms usually provide probability bounds, but have an exponential runtime. Some ran-domized schemes haw ~, a polynomial runtime, but provide only(More)
Approved for public release; distribution unlimited The views expressed in this dissertation are those of the author and do not reeect the oocial policy or position of the Department of Defense or the United States Government. Acknowledgements I would like to thank my advisor, Dr. Eugene Santos, for his insight and guidance throughout this eeort. I am also(More)
Numerous artiicial intelligence schemes and applications require, at their core, a solution to intractable computational problems, such as probabilistic reasoning. Conditions for theorems guaranteeing polynomial time algorithms for special cases, do not hold for many real-world problem instances. While there are a number of highly specialized(More)
1. The effect of bicarbonate administration on the intracellular pH of rat skeletal muscle was examined by using 31P n.m.r. 2. Bicarbonate administered intraperitoneally caused a significant intracellular acidosis in rat skeletal muscle in vivo. When the bicarbonate was administered intravenously there was no such change in the pH of the skeletal muscle. 3.(More)
We develop a new method for handling incomplete knowledge when inferencing over Bayesian Knowledge Bases(BKB) using genetic algorithms(GA). The tness function for a genetic algorithm requires that we give a score to each solution it generates. When a solution is complete, we can use the joint probability of our generated solution, but when incompleteness(More)
1. We have previously shown that the cytosolic acid concentration changes in skeletal muscle during contraction in spontaneously hypertensive rats and normotensive Wistar-Kyoto rats in vivo. We have now found that this change was unaffected by 20% inhaled CO2 or by 4,4'-di-isothiocyanostilbene-2,2'-disulphonate. This is evidence that HCO3- exchange in vivo(More)
Anytime algorithms have demonstrated their usefulness 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 improvement , both in time and system resources. While significant work has been accomplished on characterizing(More)