FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W145458542> ?p ?o ?g. }

• W145458542 abstract "Graphical models are increasingly popular tools for modeling problems involving uncertainty. They deal with uncertainty by modeling and reasoning about degrees of uncertainty explicitly based on probability theory. Practical models based on graphical models often reach the size of hundreds of variables. Although a number of ingenious inference algorithms have been developed, the problem of exact belief updating in graphical models is NP-hard. Approximate inference schemes may often be the only feasible alternative for large and complex models. The family of stochastic sampling algorithms is a promising subclass of approximate algorithms. Since previous stochastic sampling algorithms cannot converge to reasonable estimates of the posterior probabilities within a reasonable amount of time, in cases with very unlikely evidence, we cannot use the results. This thesis addresses this problem by proposing some new sampling algorithms to do the approximate inference. First, an adaptive importance sampling algorithm for Bayesian networks, AIS-BN, was developed. It shows promising convergence rates even under extreme conditions and seems to outperform the existing sampling algorithms consistently. Second, some tighter stopping rules were developed. Based on these stopping rules, two distribution-independent algorithms (the SA-μ and SA-σ algorithms) to estimate the mean of bounded random variables, one with and one without the knowledge of variance, were proposed. These two algorithms guarantee that the estimate is within the desired precision. Third, confidence inference, which can be used to calculate the probabilistic accuracy of estimate, was discussed. The new developed AISBN-μ and AISBN-σ algorithms have the attractive property of not only allowing the user to know whether approximating an inference requires a prohibitive amount of computation after a number of samples are obtained, but also allowing the user to try to avoid the prohibitive amount of computation using some other heuristic methods in AIS-BN when it happens. Last, in cases with likely evidence or with good importance sampling functions, Latin hypercube sampling and quasi-Monte Carlo methods to improve most of the current available algorithms further were developed. The experimental results show that these two methods can significantly improve the performance of sampling algorithms." @default.
• W145458542 created "2016-06-24" @default.
• W145458542 creator A5003018885 @default.
• W145458542 creator A5005313203 @default.
• W145458542 date "2001-01-01" @default.
• W145458542 modified "2023-09-28" @default.
• W145458542 title "Efficient stochastic sampling algorithms for bayesian networks" @default.
• W145458542 cites W1495157997 @default.
• W145458542 cites W1505477995 @default.
• W145458542 cites W1510302714 @default.
• W145458542 cites W1520268336 @default.
• W145458542 cites W1521447691 @default.
• W145458542 cites W1524583996 @default.
• W145458542 cites W1533955387 @default.
• W145458542 cites W1543400885 @default.
• W145458542 cites W1546613574 @default.
• W145458542 cites W1549942210 @default.
• W145458542 cites W1554807875 @default.
• W145458542 cites W1561981064 @default.
• W145458542 cites W1563726162 @default.
• W145458542 cites W1564001778 @default.
• W145458542 cites W1573686114 @default.
• W145458542 cites W1578411263 @default.
• W145458542 cites W1593793857 @default.
• W145458542 cites W1615201606 @default.
• W145458542 cites W1630331242 @default.
• W145458542 cites W1675231743 @default.
• W145458542 cites W1721794304 @default.
• W145458542 cites W1756022602 @default.
• W145458542 cites W180577587 @default.
• W145458542 cites W1919989817 @default.
• W145458542 cites W1928061009 @default.
• W145458542 cites W1969481231 @default.
• W145458542 cites W1979446688 @default.
• W145458542 cites W1980452149 @default.
• W145458542 cites W1981430627 @default.
• W145458542 cites W1986130900 @default.
• W145458542 cites W1986713664 @default.
• W145458542 cites W1987051875 @default.
• W145458542 cites W1992068214 @default.
• W145458542 cites W1994853484 @default.
• W145458542 cites W1999321705 @default.
• W145458542 cites W1999432334 @default.
• W145458542 cites W1999901857 @default.
• W145458542 cites W2008524883 @default.
• W145458542 cites W2008740515 @default.
• W145458542 cites W2012922531 @default.
• W145458542 cites W2015616421 @default.
• W145458542 cites W2017001763 @default.
• W145458542 cites W2020999234 @default.
• W145458542 cites W2033699249 @default.
• W145458542 cites W2038669746 @default.
• W145458542 cites W2044973463 @default.
• W145458542 cites W2065412317 @default.
• W145458542 cites W2073941369 @default.
• W145458542 cites W2076350013 @default.
• W145458542 cites W2081422582 @default.
• W145458542 cites W2091082553 @default.
• W145458542 cites W2092720120 @default.
• W145458542 cites W2093625674 @default.
• W145458542 cites W2093976733 @default.
• W145458542 cites W2094500233 @default.
• W145458542 cites W2102251435 @default.
• W145458542 cites W2103012681 @default.
• W145458542 cites W2110930696 @default.
• W145458542 cites W2111950130 @default.
• W145458542 cites W2112320508 @default.
• W145458542 cites W2114909350 @default.
• W145458542 cites W2126163471 @default.
• W145458542 cites W2126653228 @default.
• W145458542 cites W2128023014 @default.
• W145458542 cites W2129031807 @default.
• W145458542 cites W2130178369 @default.
• W145458542 cites W2130416410 @default.
• W145458542 cites W2134700719 @default.
• W145458542 cites W2143075689 @default.
• W145458542 cites W2146261346 @default.
• W145458542 cites W2150218618 @default.
• W145458542 cites W2152190468 @default.
• W145458542 cites W2152898342 @default.
• W145458542 cites W2159080219 @default.
• W145458542 cites W2159082327 @default.
• W145458542 cites W2160208155 @default.
• W145458542 cites W2163238526 @default.
• W145458542 cites W2163417850 @default.
• W145458542 cites W2173497437 @default.
• W145458542 cites W2182771685 @default.
• W145458542 cites W225736130 @default.
• W145458542 cites W3135037104 @default.
• W145458542 cites W32516901 @default.
• W145458542 hasPublicationYear "2001" @default.
• W145458542 type Work @default.
• W145458542 sameAs 145458542 @default.
• W145458542 citedByCount "13" @default.
• W145458542 countsByYear W1454585422016 @default.
• W145458542 crossrefType "journal-article" @default.
• W145458542 hasAuthorship W145458542A5003018885 @default.
• W145458542 hasAuthorship W145458542A5005313203 @default.
• W145458542 hasConcept C105795698 @default.
• W145458542 hasConcept C106131492 @default.

• next