Title :
Bayesian in principle, but not always in practice
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., California Univ., Berkeley, CA, USA
Abstract :
Before considering the whys and wherefores of Bayesian reasoning in practice, the author considers the different uses of the term and the role that Bayesian reasoning should play for the scientific community. Probabilistic reasoning or modeling, together with its partner, decision theory, is a key theoretical tool for addressing uncertainty in intelligent systems of all kinds. Probabilistic reasoning comes in many shapes and sizes, and does not necessarily imply the use of Bayesian methods proper, as the author discusses
Keywords :
Bayes methods; inference mechanisms; probability; statistical analysis; uncertainty handling; Bayesian reasoning; decision theory; intelligent systems; probabilistic modeling; probabilistic reasoning; scientific community; uncertainty handling; Bayesian methods; Decision theory; Graphical models; Integrated circuit modeling; Intelligent systems; Machine learning; Neural networks; Shape; Statistics; Uncertainty;
Journal_Title :
IEEE Expert
