G. Cooper, The computational complexity of Bayesian inference using Bayesian belief networks, Artificial Intelligence 42 (2) (1990) 393--405.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....can impose noticeable delays for computation in decisionsupport systems depending on the platform and the problem being solved. Probabilistic E. Horvitz Artificial Intelligence 126 (2001) 159 196 185 inference in general graphical models like Bayesian networks and influence diagrams is NP hard [19,20]. The computation of EVI, even for the case of the greedy analysis, requires, for each piece of unobserved evidence, probabilistic inference about the outcome of seeing the spectrum of alternate values should that observation be carried out. Multiple computations for each potential test or ....

G. Cooper, The computational complexity of Bayesian inference using Bayesian belief networks, Artificial Intelligence 42 (2) (1990) 393--405.

....computation of the net expected value of information (NEVI) can impose noticeable delays for computation in decision support systems depending on the platform and the problem being solved. Probabilistic inference in general graphical models like Bayesian networks and influence diagrams is NP hard (Cooper 1990; Dagum Luby 1993) The computation of NEVI, even for the case of the greedy analysis, requires, for each piece of unobserved evidence, probabilistic inference about the outcome of seeing the spectrum of alternate values should that observation be carried out. Thus, multiple computations for ....

Cooper, G. 1990. The computational complexity of bayesian inference using bayesian belief networks.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC