Treatment of uncertainty in artificial intelligenceThe present assessment of the development status of research efforts concerned with AI reasoning under conditions of uncertainty emphasizes the importance of appropriateness in the approach selected for both the epistemic and the computational levels. At the former level, attention is given to the form of uncertainty-representation and the fidelity of its reflection of actual problems' uncertainties; at the latter level, such issues as the availability of the requisite information and the complexity of the reasoning process must be considered. The tradeoff between these levels must always be the focus of AI system-developers' attention.
Document ID
19890043715
Acquisition Source
Legacy CDMS
Document Type
Conference Proceedings
Authors
Berenji, Hamid R. (NASA Ames Research Center; Heer Associates, Inc. Moffett Field, CA, United States)