training conditional random fields using virtual evidence boosting

training conditional random fields using virtual evidence boosting

ID：34414996

大小：142.61 KB

頁數(shù)：6頁

時間：2019-03-05

training conditional random fields using virtual evidence boosting_第1頁
training conditional random fields using virtual evidence boosting_第2頁
training conditional random fields using virtual evidence boosting_第3頁
training conditional random fields using virtual evidence boosting_第4頁
training conditional random fields using virtual evidence boosting_第5頁
資源描述：

《training conditional random fields using virtual evidence boosting》由會員上傳分享，免費在線閱讀，更多相關內容在教育資源-天天文庫。

1、TrainingConditionalRandomFieldsusingVirtualEvidenceBoostingLinLiaoTanzeemChoudhury?DieterFoxHenryKautzUniversityofWashington?IntelResearchDepartmentofComputerScience&Engineering1100NE45thSt.Seattle,WA98195Seattle,WA98105Abstracteraldomains.However,nogeneralguidancehasbeengivenonwhenMPL

2、canbesafelyused,andindeedMPLhasbeenWhileconditionalrandom?elds(CRFs)havebeenobservedtoover-estimatethedependencyparametersinsomeappliedsuccessfullyinavarietyofdomains,theirexperiments[GeyerandThompson,1992].trainingremainsachallengingtask.Inthispaper,Inaddition,neitherMLnorMPLperformsf

3、eatureselec-weintroduceanoveltrainingmethodforCRFs,tionexplicitly,andneitherofthemisabletoadequatelyhan-calledvirtualevidenceboosting,whichsimulta-dlecontinuousobservations.Theselimitationsmakethemneouslyperformsfeatureselectionandparameterunsuitableforsometasks,suchasactivityrecogniti

4、onbasedestimation.Toachievethis,weextendstandardonrealsensordataandidentifyingthesetoffeaturesthatboostingtohandlevirtualevidence,whereanob-aremostusefulforclassi?cation.Alternatively,boostinghasservationcanbespeci?edasadistributionratherbeensuccessfullyusedforfeatureselectioninthecont

5、extofthanasinglenumber.Thisextensionallowsustoclassi?cationproblems[ViolaandJones,2002].However,itsdevelopauni?edframeworkforlearningbothlocalapplicationtorelationaldataremainsanunsolvedproblemandcompatibilityfeaturesinCRFs.Inexperimentssinceitassumestheindependenceofhiddenlabels.onsyn

6、theticdataaswellasrealactivityclassi?-Inthispaper,weshowhowtoseamlesslyintegrateboost-cationproblems,ournewtrainingalgorithmout-ingandCRFtraining,therebycombiningthecapabilitiesofperformsothertrainingapproachesincludingmax-bothparadigms.Theintegrationisachievedbycuttingaimumlikelihood,

7、maximumpseudo-likelihood,andCRFintoindividualpatches,asdoneinMPL,andusingthesethemostrecentboostedrandom?elds.patchesastraininginstancesforboosting.ThekeydifferencetoMPL,however,isthatinourframeworktheneighborlabels1Introductionarenottreatedasobserved,butasvirtualevidencesorbeliefs.T

當前文檔最多預覽五頁，下載文檔查看全文

此文檔下載收益歸作者所有

當前文檔最多預覽五頁，下載文檔查看全文
溫馨提示：
1. 部分包含數(shù)學公式或PPT動畫的文件，查看預覽時可能會顯示錯亂或異常，文件下載后無此問題，請放心下載。
2. 本文檔由用戶上傳，版權歸屬用戶，天天文庫負責整理代發(fā)布。如果您對本文檔版權有爭議請及時聯(lián)系客服。
3. 下載前請仔細閱讀文檔內容，確認文檔內容符合您的需求后進行下載，若出現(xiàn)內容與標題不符可向本站投訴處理。
4. 下載文檔時可能由于網絡波動等原因無法下載或下載錯誤，付費完成后未能成功下載的用戶請聯(lián)系客服處理。