Statistical Learning Theory

Statistical Learning Theory

ID:40702186

大小:689.38 KB

页数:105页

时间:2019-08-06

Statistical Learning Theory_第1页
Statistical Learning Theory_第2页
Statistical Learning Theory_第3页
Statistical Learning Theory_第4页
Statistical Learning Theory_第5页
资源描述:

《Statistical Learning Theory》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、CS229T/STAT231:StatisticalLearningTheory(Winter2014)PercyLiangLastupdatedWedFeb12201409:24Theselecturenoteswillbeupdatedperiodicallyasthecoursegoeson.Pleaseletusknowwhenyou ndtypos.SectionA.1describesthebasicnotationandde nitions;SectionA.2describessomebasictheorems.LecturesThenotesareorganiz

2、edaccordingtotopic.However,thelecturesthemselveswilloftenspillacrosstopics,sothematerialactuallycoveredineachlectureisexplicitlydelimitedandlinkedbelow.Lecture1/6:Overview,onlinelearningLecture1/8:OnlinelearningLecture1/13:OnlinelearningLecture1/15:OnlinelearningLecture1/20:Noclass(Marti

3、nLutherKingDay)Lecture1/22:OnlinelearningLecture1/27:KernelmethodsLecture1/29:KernelmethodsLecture2/3:KernelmethodsLecture2/5:UniformconvergenceLecture2/10:UniformconvergenceLecture2/12:UniformconvergenceLecture2/17:Noclass(PresidentsDay)Lecture2/19:UniformconvergenceLecture2/24:Dir

4、ectanalysisLecture2/26:DirectanalysisLecture3/3:Spectralmethods1Lecture3/5:SpectralmethodsLecture3/10:SpectralmethodsLecture3/12:Finalprojectpresentations21Overview[beginlecture1/6](1/6)1.1Whatisthiscourseabout?Machinelearninghasbecomeanindispensiblepartofmanyapplicationareas,rangingfro

5、mscience(biology,neuroscience,psychology,astronomy,etc.)toengineering(nat-urallanguageprocessing,computervision,robotics,etc.).Butmachinelearningisnotonecoherentapproach,butratherconsistsofadazzlingarrayofseeminglydisparateframeworksandparadigmsspanningclassi cation,regression,clustering,matr

6、ixfac-torization,graphicalmodels,probabilisticprogramming,etc.Thiscoursetriestouncoverthecommonstatisticalprinciplesunderlyingthisdiversearrayoftechniques,usingtheoreticalanalysisasthemaintoolforunderstanding.Wewilltrytoanswertwoquestions:{Whendolearningalgorithmswork(orfail)?Anyonewithhands

7、-onmachinelearningexperiencewillprobablyknowthatlearningalgorithmsrarelyworkthe rsttime.Whentheydon't,adeepertheoreticalunderstandingcano eranewperspectiveandcanaidintroubleshooting.{Howdowemakelearningalgorithmsbetter?Theoreticalanalysesofte

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。