Prediction, Learning, and Games .pdf

Prediction, Learning, and Games .pdf

ID:34820188

大小:5.67 MB

页数:407页

时间:2019-03-11

Prediction, Learning, and Games .pdf_第1页
Prediction, Learning, and Games .pdf_第2页
Prediction, Learning, and Games .pdf_第3页
Prediction, Learning, and Games .pdf_第4页
Prediction, Learning, and Games .pdf_第5页
Prediction, Learning, and Games .pdf_第6页
Prediction, Learning, and Games .pdf_第7页
Prediction, Learning, and Games .pdf_第8页
Prediction, Learning, and Games .pdf_第9页
Prediction, Learning, and Games .pdf_第10页
资源描述:

《Prediction, Learning, and Games .pdf》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、Prediction,Learning,andGamesThisimportantnewtextandreferenceforresearchersandstudentsinmachinelearning,gametheory,statistics,andinformationtheoryoffersthefirstcomprehensivetreatmentoftheproblemofpredictingindividualsequences.Unlikestandardstatisticalapproachestoforecasting,predic-tionofin

2、dividualsequencesdoesnotimposeanyprobabilisticassump-tiononthedata-generatingmechanism.Yet,predictionalgorithmscanbeconstructedthatworkwellforallpossiblesequences,inthesensethattheirperformanceisalwaysnearlyasgoodasthebestforecastingstrategyinagivenreferenceclass.Thecentralthemeisthemode

3、lofpredictionusingexpertadvice,ageneralframeworkwithinwhichmanyrelatedproblemscanbecastanddiscussed.Repeatedgameplaying,adaptivedatacompression,sequentialinvestmentinthestockmarket,sequentialpatternanalysis,andseveralotherproblemsareviewedasinstancesoftheexperts’frameworkandanalyzedfroma

4、commonnonstochasticstandpointthatoftenrevealsnewandintriguingconnections.Oldandnewforecastingmethodsaredescribedinamathematicallyprecisewayinordertocharacterizetheirtheoreticallimitationsandpossibilities.NicoloCesa-BianchiisProfessorofComputerScienceattheUniversity`ofMilan,Italy.Hisresea

5、rchinterestsincludelearningtheory,patternanalysis,andworst-caseanalysisofalgorithms.HeisanactioneditorofthejournalMachineLearning.GaborLugosihasbeenworkingonvariousproblemsinpatternclassi´fi-cation,nonparametricstatistics,statisticallearningtheory,gametheory,probability,andinformationtheo

6、ry.HeiscoauthorofthemonographsAProbabilisticTheoryofPatternRecognitionandCombinatorialMeth-odsofDensityEstimation.Hehasbeenanassociateeditorofvariousjournals,includingtheIEEETransactionsofInformationTheory,Test,ESAIM:ProbabilityandStatistics,andStatisticsandDecisions.Prediction,Learning,

7、andGamesNICOLOCESA-BIANCHI`UniversitadegliStudidiMilano`GABORLUGOSI´UniversitatPompeuFabra,BarcelonacambridgeuniversitypressCambridge,NewYork,Melbourne,Madrid,CapeTown,Singapore,SãoPauloCambridgeUniversityPressTheEdinburghBuilding,Cambridgecb22ru,UKPublishedintheUnitedSta

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。