返回
approximating max sum product problems using multiplicative error bounds what s the h in h likelihood a holy grail or an achilles heel

approximating max sum product problems using multiplicative error bounds what s the h in h likelihood a holy grail or an achilles heel

ID：7283627

大小：467.96 KB

页数：35页

时间：2018-02-10

approximating max sum product problems using multiplicative error bounds what s the h in h likelihood a holy grail or an achilles heel_第1页
approximating max sum product problems using multiplicative error bounds what s the h in h likelihood a holy grail or an achilles heel_第2页
approximating max sum product problems using multiplicative error bounds what s the h in h likelihood a holy grail or an achilles heel_第3页
approximating max sum product problems using multiplicative error bounds what s the h in h likelihood a holy grail or an achilles heel_第4页
approximating max sum product problems using multiplicative error bounds what s the h in h likelihood a holy grail or an achilles heel_第5页
资源描述：

《approximating max sum product problems using multiplicative error bounds what s the h in h likelihood a holy grail or an achilles heel》由会员上传分享，免费在线阅读，更多相关内容在工程资料-天天文库。

1、ApproximatingMax‐Sum‐ProductProblemsusingMultiplicativeErrorBoundsUniversityPressScholarshipOnlineOxfordScholarshipOnlineBayesianStatistics9JoséM.Bernardo,M.J.Bayarri,JamesO.Berger,A.P.Dawid,DavidHeckerman,AdrianF.M.Smith,andMikeWestPrintpublicationdate:2011PrintISBN-13:9780199694587PublishedtoOxfor

2、dScholarshipOnline:January2012DOI:10.1093/acprof:oso/9780199694587.001.0001ApproximatingMaxSumProductProblemsusingMultiplicativeErrorBoundsChristopherMeekYdoWexlerDOI:10.1093/acprof:oso/9780199694587.003.0015AbstractandKeywordsWedescribetheMultiplicativeApproximationScheme(MAS)forapproximateinferenc

3、einmultiplicativemodels.WeapplythisschemetodeveloptheDynaDecompapproximationalgorithm.Thisalgorithmcanbeusedtoobtainboundedapproximationsforvarioustypesofmax‐sum‐productproblemsincludingthecomputationofthelogprobabilityofevidence,thelog‐partitionfunction,MostProbableExplanation(MPE)andmaximumaposter

4、ioriprobability(MAP)inferenceproblems.Wedemonstratethatthisalgorithmyieldsboundedapproximationssuperiortoexistingmethodsusingavarietyoflargegraphicalmodels.Keywords:Approxmatenference,graphcamodes,max‐sum‐product,sum‐productSummaryWedescribetheMultiplicativeApproximationScheme(MAS)forapproximateinfe

5、renceinmultiplicativemodels.WeapplythisschemetodeveloptheDynaDecompapproximationalgorithm.Thisalgorithmcanbeusedtoobtainboundedapproximationsforvarioustypesofmax‐sum‐productproblemsincludingthecomputationofthelogprobabilityofevidence,thePage1of35ApproximatingMax‐Sum‐ProductProblemsusingMultiplicativ

6、eErrorBoundslog‐partitionfunction,MostProbableExplanation(MPE)andmaximumaposterioriprobability(MAP)inferenceproblems.Wedemonstratethatthisalgorithmyieldsboundedapproximationssuperiortoexistingmethodsusingavarietyoflargegraphicalmodels.KeywordsandPhrases:APPROXIMATEINFERENCE,GRAPHICALMODELS,MAX‐SUM‐P

7、RODUCT,SUM‐PRODUCT1.IntroductionProbabilisticgraphicalmodelshavegainedpopularityinrecentdecadesduetotheirintuitiverepresentationandbecauseoftheexistenceofalgorithmsforansweringprobabilisticinferencepr

当前文档最多预览五页，下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页，下载文档查看全文
温馨提示：
1. 部分包含数学公式或PPT动画的文件，查看预览时可能会显示错乱或异常，文件下载后无此问题，请放心下载。
2. 本文档由用户上传，版权归属用户，天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容，确认文档内容符合您的需求后进行下载，若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误，付费完成后未能成功下载的用户请联系客服处理。