
This Article  
 
Share  
Bibliographic References  
Add to:  
Digg Furl Spurl Blink Simpy Del.icio.us Y!MyWeb  
Search  
 
ASCII Text  x  
D. Heckerman, E. Horvitz, B. Middleton, "An Approximate Nonmyopic Computation for Value of Information," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 15, no. 3, pp. 292298, March, 1993.  
BibTex  x  
@article{ 10.1109/34.204912, author = {D. Heckerman and E. Horvitz and B. Middleton}, title = {An Approximate Nonmyopic Computation for Value of Information}, journal ={IEEE Transactions on Pattern Analysis and Machine Intelligence}, volume = {15}, number = {3}, issn = {01628828}, year = {1993}, pages = {292298}, doi = {http://doi.ieeecomputersociety.org/10.1109/34.204912}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, }  
RefWorks Procite/RefMan/Endnote  x  
TY  JOUR JO  IEEE Transactions on Pattern Analysis and Machine Intelligence TI  An Approximate Nonmyopic Computation for Value of Information IS  3 SN  01628828 SP292 EP298 EPD  292298 A1  D. Heckerman, A1  E. Horvitz, A1  B. Middleton, PY  1993 KW  belief networks; decision theory; information value; probability; approximate nonmyopic computation; belief maintenance; decision theory; information theory; probability VL  15 JA  IEEE Transactions on Pattern Analysis and Machine Intelligence ER   
It is argued that decision analysts and expertsystem designers have avoided the intractability of exact computation of the value of information by relying on a myopic assumption that only one additional test will be performed, even when there is an opportunity to make large number of observations. An alternative to the myopic analysis is presented. In particular, an approximate method for computing the value of information of a set of tests, which exploits the statistical properties of large samples, is given. The approximation is linear in the number of tests, in contrast with the exact computation, which is exponential in the number of tests. The approach is not as general as in a complete nonmyopic analysis, in which all possible sequences of observations are considered. In addition, the approximation is limited to specific classes of dependencies among evidence and to binary hypothesis and decision variables. Nonetheless, as demonstrated with a simple application, the approach can offer an improvement over the myopic analysis.
[1] G. A. Gorry and G. O. Barnett, "Experience with a model of sequential diagnosis,"Comput. Biomed. Res., vol. 1, pp. 490507, 1968.
[2] G. A. Gorry, J. P. Kassirer, A. Essig, and W. B. Schwartz, "Decision analysis as the basis for computeraided management of acute renal failure,"Amer. J. Med., vol. 55, pp. 473484, 1973.
[3] D. E. Heckerman, E. J. Horvitz, and B. N. Nathwani, "Pathfinder research directions," Tech. Rep. KSL8964, Med. Comput. Sci. Group, Section on Med. Informatics, Stanford Univ., Stanford, CA, Oct. 1985.
[4] D. E. Heckerman,Probabilistic Similarity Networks. Cambridge, MA: MIT Press, 1991.
[5] D. Heckerman, E. Horvitz, and B. N. Nathwani, "Toward normative expert systems: Part I. The Pathfinder project,"Methods Inform. Med., vol. 31, pp. 90105, 1992.
[6] B. N. Nathwani, D. E. Heckerman, E. J. Horvitz, and T. L. Lincoln, "Integrated expert systems and videodisc in surgical pathology: An overview,"Human Pathol., vol. 21, pp. 1127, 1990.
[7] R. A. Howard, "Value of information lotteries,"IEEE Trans. Syst. Sci. Cybern., vol. SSC3, no. 1, pp. 5460, 1967.
[8] J. Pearl,Probabilistic Reasoning in Intelligent Systems. San Mateo, CA: Morgan Kaufmann, 1988.
[9] P. Billingsley, "Dependent variables," inConvergence of Probability Measures. New York: Wiley, 1968, ch. 4.
[10] G. F. Cooper and E. Herskovits, "A Bayesian method for constructing Bayesian belief networks from databases," inProc. Seventh Conf. Uncertainty Artificial Intell., 1991, pp. 8694.
[11] B. J. McNeil, S. G. Pauker, H. C. Sox, and A. Tversky, "On the elicitation of preferences for alternative therapies,"New Eng. J. Med., vol. 306, pp. 12591262, 1982.
[12] R. A. Howard, "On making life and death decisions," inSocietal Risk Assessment(R. C. Schwing and W. A. Albers, Jr., Eds). New York: Plenum, 1980, pp. 89113.
[13] D. E. Heckerman and E. J. Horvitz, "Problem formulation as the reduction of a decision model," inProc. Sixth Conf. Uncertainty Artificial Intell.(Boston, MA), July 1990, pp. 8289; also in P. Bonissone, M. Henrion, L. Kanal, and J. Lemmer, Eds.Uncertainty in Artificial Intelligence 6. New York: NorthHolland, 1990, pp. 159170.