Reasoning About Beliefs and Actions Under Computational Resource Constraints

From Tetherless World Wiki

Jump to: navigation, search

Citation: Eric Horvitz. (1987) Reasoning About Beliefs and Actions Under Computational Resource Constraints. In Proceedings of the Third Workshop on Uncertainty in Artificial Intelligence, 429-444,1987.

Publication inproceedings ( Edit )
type InProceedings
bibtype inproceedings
Bibtex basics
author Eric Horvitz
title Reasoning About Beliefs and Actions Under Computational Resource Constraints
booktitle Proceedings of the Third Workshop on Uncertainty in Artificial Intelligence
pages 429-444
address Seattle, Washington
year 1987
Bibtex more
note Also in L. Kanal, T. Levitt, and J. Lemmer, ed., Uncertainty in Artificial Intelligence 3, Elsevier, 1989, pps. 301-324. http://www.research.microsoft.com/research/dtg/horvitz/u87.htm.
Access Paper
abstract It is important for computers systems solving problems in real-world environments to have the ability to respond flexibly to wide variations in the availability of resources needed for inference. For example, the time available for computation may vary greatly depending on the criticality of the situation. The proposed research is focused on handling resource problems through endowing computational systems with knowledge about the value of computation with compex decision-making contexts. The intent is to develop techniques for reasoning about the optimal allocation of computational resources in solving diverse problems. Methods for considering resource limitations and inference tradeoffs will be especially crucial in computer systems that can provide decision support and automated control in complex and critical aerospace application areas. The research will include (1) the assignment of context-sensitive degrees of value to reasoning strategy performance, (2) the study of relationships among limited resources, (3) characterization of the behavior of uncertain reasoning strategies in varying problem settings, and (4) the development of techniques for reasoning about tradeoffs in order to optimize the value of inference to users.

KSL Technical Report ID: KSL-86-55
Facts about Reasoning About Beliefs and Actions Under Computational Resource ConstraintsRDF feed
Abstract It is important for computers systems solv It is important for computers systems solving problems in real-world environments to have the ability to respond flexibly to wide variations in the availability of resources needed for inference. For example, the time available for computation may vary greatly depending on the criticality of the situation. The proposed research is focused on handling resource problems through endowing computational systems with knowledge about the value of computation with compex decision-making contexts. The intent is to develop techniques for reasoning about the optimal allocation of computational resources in solving diverse problems. Methods for considering resource limitations and inference tradeoffs will be especially crucial in computer systems that can provide decision support and automated control in complex and critical aerospace application areas. The research will include (1) the assignment of context-sensitive degrees of value to reasoning strategy performance, (2) the study of relationships among limited resources, (3) characterization of the behavior of uncertain reasoning strategies in varying problem settings, and (4) the development of techniques for reasoning about tradeoffs in order to optimize the value of inference to users. optimize the value of inference to users.
Address Seattle, Washington  +
Author Eric Horvitz  +
Bibtype inproceedings  +
Booktitle Proceedings of the Third Workshop on Uncertainty in Artificial Intelligence  +
Has author Eric Horvitz  +
Has identifier KSL-86-55  +
Has publishing details 429-444,1987  +
Has title Reasoning About Beliefs and Actions Under Computational Resource Constraints  +
Has where published Proceedings of the Third Workshop on Uncertainty in Artificial Intelligence  +
Has year 1987  +
Ksl tr id KSL-86-55  +
Note Also in L. Kanal, T. Levitt, and J. Lemmer, ed., Uncertainty in Artificial Intelligence 3, Elsevier, 1989, pps. 301-324. http://www.research.microsoft.com/research/dtg/horvitz/u87.htm.
Pages 429-444  +
Process note GOOGLE  +
Title Reasoning About Beliefs and Actions Under Computational Resource Constraints  +
Year 1987  +
Personal tools