Machine Learning with Python: from Linear Models to Deep ...

An in-depth introduction to the field of machine learning, from linear models to deep learning and reinforcement learning, through hands-on Python projects. Thereisonesessionavailable:163,413alreadyenrolled!StartedMay16EndsSep10EnrollIwouldliketoreceiveemailfromMITxandlearnaboutotherofferingsrelatedtoMachineLearningwithPython:fromLinearModelstoDeepLearning.AboutthiscourseWhatyou'lllearnInstructorsFrequentlyAskedQuestionsWaystotakethiscourseedXforbusinessEstimated15weeks10–14hoursperweekInstructor-pacedInstructor-ledonacoursescheduleFreeOptionalupgradeavailableThereisonesessionavailable:163,413alreadyenrolled!Afteracoursesessionends,itwillbearchivedOpensinanewtab.StartedMay16EndsSep10EnrollIwouldliketoreceiveemailfromMITxandlearnaboutotherofferingsrelatedtoMachineLearningwithPython:fromLinearModelstoDeepLearning.MachineLearningwithPython:fromLinearModelstoDeepLearningAboutAboutWhatyou'lllearnSyllabusInstructorsFAQsWaystoenrollEnrollnowStartedMay16AboutthiscourseSkipAboutthiscourseIfyouhavespecificquestionsaboutthiscourse,[email protected]. Machinelearningmethodsarecommonlyusedacrossengineeringandsciences,fromcomputersystemstophysics.Moreover,commercialsitessuchassearchengines,recommendersystems(e.g.,Netflix,Amazon),advertisers,andfinancialinstitutionsemploymachinelearningalgorithmsforcontentrecommendation,predictingcustomerbehavior,compliance,orrisk. Asadiscipline,machinelearningtriestodesignandunderstandcomputerprogramsthatlearnfromexperienceforthepurposeofpredictionorcontrol. Inthiscourse,studentswilllearnaboutprinciplesandalgorithmsforturningtrainingdataintoeffectiveautomatedpredictions.Wewillcover: Representation,over-fitting,regularization,generalization,VCdimension; Clustering,classification,recommenderproblems,probabilisticmodeling,reinforcementlearning; On-linealgorithms,supportvectormachines,andneuralnetworks/deeplearning. StudentswillimplementandexperimentwiththealgorithmsinseveralPythonprojectsdesignedfordifferentpracticalapplications. ThiscourseispartoftheMITxMicroMastersPrograminStatisticsandDataScience.Mastertheskillsneededtobeaninformedandeffectivepractitionerofdatascience.YouwillcompletethiscourseandthreeothersfromMITx,atasimilarpaceandlevelofrigorasanon-campuscourseatMIT,andthentakeavirtually-proctoredexamtoearnyourMicroMasters,anacademiccredentialthatwilldemonstrateyourproficiencyindatascienceoraccelerateyourpathtowardsanMITPhDoraMaster'satotheruniversities.Tolearnmoreaboutthisprogram,pleasevisithttps://micromasters.mit.edu/ds/. Pleasenote:edXInc.hasrecentlyenteredintoanagreementtotransfertheedXplatformto2U,Inc.,whichwillcontinuetoruntheplatformthereafter.Thesalewillnotaffectyourcourseenrollment,coursefeesorchangeyourcourseexperienceforthisoffering.ItispossiblethattheclosingofthesaleandthetransferoftheedXplatformmaybeeffectuatedsometimeintheFallwhilethiscourseisrunning.PleasebeawarethattherecouldbechangestotheedXplatformPrivacyPolicyorTermsofServiceaftertheclosingofthesale.However,2Uhascommittedtopreservingrobustprivacyofindividualdataforalllearnerswhousetheplatform.FormoreinformationseetheedXHelpCenter.ShowmoreAtaglanceInstitution:MITxSubject:ComputerScienceLevel:AdvancedPrerequisites: 6.00.1xorproficiencyinPythonprogramming 6.431xorequivalentprobabilitytheorycourse College-levelsingleandmulti-variablecalculus Vectorsandmatrices Language:EnglishVideoTranscript:EnglishAssociatedprograms:MicroMasters®PrograminStatisticsandDataScienceMicroMasters®PrograminStatisticsandDataScience(Generaltrack)Whatyou'lllearnSkipWhatyou'lllearn Understandprinciplesbehindmachinelearningproblemssuchasclassification,regression,clustering,andreinforcementlearning Implementandanalyzemodelssuchaslinearmodels,kernelmachines,neuralnetworks,andgraphicalmodels Choosesuitablemodelsfordifferentapplications Implementandorganizemachinelearningprojects,fromtraining,validation,parametertuning,tofeatureengineering. ShowmoreSyllabusSkipSyllabusLectures: Introduction Linearclassifiers,separability,perceptronalgorithm Maximummarginhyperplane,loss,regularization Stochasticgradientdescent,over-fitting,generalization Linearregression Recommenderproblems,collaborativefiltering Non-linearclassification,kernels Learningfeatures,Neuralnetworks Deeplearning,backpropagation Recurrentneuralnetworks Recurrentneuralnetworks Generalization,complexity,VC-dimension Unsupervisedlearning:clustering Generativemodels,mixtures MixturesandtheEMalgorithm Learningtocontrol:Reinforcementlearning Reinforcementlearningcontinued Applications:NaturalLanguageProcessing Projects: AutomaticReviewAnalyzer DigitRecognitionwithNeuralNetworks ReinforcementLearning ShowmoreAbouttheinstructorsFrequentlyAskedQuestionsSkipFrequentlyAskedQuestionsShouldyouhavefurtherinquiries,pleasegotohttps://micromasters.mit.edu/ds/andusethe"Contactus"button.ShowmoreWhocantakethiscourse?Unfortunately,learnersresidinginoneormoreofthefollowingcountriesorregionswillnotbeabletoregisterforthiscourse:Iran,CubaandtheCrimearegionofUkraine.WhileedXhassoughtlicensesfromtheU.S.OfficeofForeignAssetsControl(OFAC)toofferourcoursestolearnersinthesecountriesandregions,thelicenseswehavereceivedarenotbroadenoughtoallowustoofferthiscourseinalllocations.edXtrulyregretsthatU.S.sanctionspreventusfromofferingallofourcoursestoeveryone,nomatterwheretheylive.WaystotakethiscourseChooseyourpathwhenyouenroll.EnrollnowStartedMay16VerifiedTrackAuditTrackPrice$300USDFreeAccesstocoursematerialsUnlimitedLimitedExpiresonAug29WorldclassinstitutionsanduniversitiesedXsupportShareablecertificateuponcompletionGradedassignmentsandexamsReadourFAQsinanewtababoutfrequentlyaskedquestionsonthesetracks.Interestedinthiscourseforyourbusinessorteam?Trainyouremployeesinthemostin-demandtopics,withedXforBusiness.PurchasenowRequestinformation