资源描述：
《机器学习课件.ppt》由会员上传分享，免费在线阅读，更多相关内容在PPT专区-天天文库。
1、MachineLearningLecture1Introduction1MotivatingProblemsHandwrittenCharacterRecognition2MotivatingProblemsFingerprintRecognition(e.g.,bordercontrol)3MotivatingProblemsFaceRecognition(securityaccesstobuildingsetc)4CanMachinesLearntoSolveTheseProblems?Or,tobemorepreciseCanweprogrammachinestolea
2、rntodothesetasks?5DefinitionofLearningAcomputerprogramissaidtolearnfromexperienceEwithrespecttosomeclassoftasksTandperformancemeasureP,ifitsperformanceattasksinT,asmeasuredbyP,improveswithexperienceE(Mitchell,MachineLearning,McGraw-Hill,1997)6DefinitionofLearningWhatdoesthismeanexactly?Hand
3、writingrecognitionproblemTaskT:RecognizinghandwrittencharactersPerformancemeasureP:percentofcharacterscorrectlyclassifiedTrainingexperienceE:adatabaseofhandwrittencharacterswithgivenclassifications7DesignaLearningSystemWeshallusehandwrittenCharacterrecognitionasanexampletoillustratethedesig
4、nissuesandapproaches8DesignaLearningSystemStep0:LetstreatthelearningsystemasablackboxLearningSystemZ9DesignaLearningSystemStep1:CollectTrainingExamples(Experience).Withoutexamples,oursystemwillnotlearn(so-calledlearningfromexamples)23678910DesignaLearningSystemStep2:RepresentingExperienceCh
5、oosearepresentationschemefortheexperience/examplesThesensorinputrepresentedbyann-dvector,calledthefeaturevector,X=(x1,x2,x3,…,xn)(1,1,0,1,1,1,1,1,1,1,0,0,0,0,1,1,1,1,1,0,….,1)64-dVector(1,1,1,1,1,1,1,1,1,1,0,0,1,1,1,1,1,1,1,0,….,1)64-dVector11DesignaLearningSystemStep2:RepresentingExperienc
6、eChoosearepresentationschemefortheexperience/examplesThesensorinputrepresentedbyann-dvector,calledthefeaturevector,X=(x1,x2,x3,…,xn)Torepresenttheexperience,weneedtoknowwhatXis.SoweneedacorrespondingvectorD,whichwillrecordourknowledge(experience)aboutXTheexperienceEisapairofvectorsE=(X,D)12
7、DesignaLearningSystemStep2:RepresentingExperienceChoosearepresentationschemefortheexperience/examplesTheexperienceEisapairofvectorsE=(X,D)So,whatwouldDbelike?Therearemanypossibilities.13DesignaLearningSystemStep2:RepresentingExperienceSo,whatwouldDbelike
