machine learning that matters

machine learning that matters

ID:39716899

大?。?33.21 KB

頁數(shù):6頁

時間:2019-07-10

machine learning that matters_第1頁
machine learning that matters_第2頁
machine learning that matters_第3頁
machine learning that matters_第4頁
machine learning that matters_第5頁
資源描述:

《machine learning that matters》由會員上傳分享,免費在線閱讀,更多相關(guān)內(nèi)容在學術(shù)論文-天天文庫。

1、MachineLearningthatMattersKiriL.Wagsta?kiri.l.wagstaff@jpl.nasa.govJetPropulsionLaboratory,CaliforniaInstituteofTechnology,4800OakGroveDrive,Pasadena,CA91109USAAbstracttivelysolvedspamemaildetection(Zdziarski,2005)andmachinetranslation(Koehnetal.,2003),twoMuchofcurre

2、ntmachinelearning(ML)re-problemsofglobalimport.Andsoon.searchhaslostitsconnectiontoproblemsofimporttothelargerworldofscienceandso-Andyetwestillobserveaproliferationofpublishedciety.Fromthisperspective,thereexistglar-MLpapersthatevaluatenewalgorithmsonahandfulinglimit

3、ationsinthedatasetsweinvesti-ofisolatedbenchmarkdatasets.Their“realworld”gate,themetricsweemployforevaluation,experimentsmayoperateondatathatoriginatedinandthedegreetowhichresultsarecommu-therealworld,buttheresultsarerarelycommunicatednicatedbacktotheiroriginatingdom

4、ains.backtotheorigin.Quantitativeimprovementsinper-Whatchangesareneededtohowwecon-formancearerarelyaccompaniedbyanassessmentofductresearchtoincreasetheimpactthatMLwhetherthosegainsmattertotheworldoutsideofhas?WepresentsixImpactChallengestoex-machinelearningresearch.p

5、licitlyfocusthe?eld’senergyandattention,Thisphenomenonoccursbecausethereisnoandwediscussexistingobstaclesthatmustwidespreademphasis,inthetrainingofgraduatestu-beaddressed.Weaimtoinspireongoingdis-dentresearchersorinthereviewprocessforsubmittedcussionandfocusonMLthatm

6、atters.papers,onconnectingMLadvancesbacktothelargerworld.Eventherichassortmentofapplications-drivenMLresearchoftenfailstotakethe?nalsteptotrans-1.Introductionlateresultsintoimpact.Atonetimeoranother,weallencounterafriend,Manymachinelearningproblemsarephrasedintermssp

7、ouse,parent,child,orconcernedcitizenwho,uponofanobjectivefunctiontobeoptimized.Itistimeforlearningthatweworkinmachinelearning,wondersustoaskaquestionoflargerscope:whatisthe?eld’s“What’sitgoodfor?”Thequestionmaybephrasedobjectivefunction?Doweseektomaximizeperfor-mores

8、ubtlyorelegantly,butnomatteritsform,itgetsmanceonisolateddatasets?Orcanwecharacterizeatthemotivationalunderpinningsoftheworkthatwep

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(quán)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(quán)有爭議請及時聯(lián)系客服。
3. 下載前請仔細閱讀文檔內(nèi)容,確認文檔內(nèi)容符合您的需求后進行下載,若出現(xiàn)內(nèi)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(wǎng)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。