ThedatasetiscollectedbyYonseiUniversity.Wedeployedourmobilitymonitoringsystem,namedLifeMap,tocollectmobilitydataovertwomonthsinSeoul,Korea.LifeMapusedlearningschemeproposedinfollowingpaper.Pleasereferthispaperwhenyouuseourdataset.*YohanChon,ElmurodTalipov,HyojeongShin,andHojungCha.2011.Mobilityprediction-basedsmartphoneenergyoptimizationforeverydaylocationmonitoring.InProceedingsofthe9thACMConferenceonEmbeddedNetworkedSensorSystems(SenSys'11).ACM,NewYork,NY,USA,82-95.Visitourhomepageformoreinformation(http://lifemap.yonsei.ac.kr).

关于凸分析和最优化的经典教材，机器学习重要参考书。

优化觅食算法（OFA2016年），代码包含论文中所述的觅食操作。

教学优化（teachinglearningbasedoptimization）相关学位论文，

anintroductiontooptimization4theditionsolutionmanual.pdf最优化导论（第四版）课后习题答案

CONTENTPART1BASICSOFINFERENCEOVERNETWORKSCHAPTER1AsynchronousAdaptiveNetworksCHAPTER2EstimationandDetectionOverAdaptiveNetworksCHAPTER3MultitaskLearningOverAdaptiveNetworksWithGroupingCHAPTER4BayesianApproachtoCollaborativeInferenceinNetworksCHAPTER5MultiagentDistributedOptimizationCHAPTER6DistributedKalmanandParticleFilteringCHAPTER7GameTheoreticLearningPART2SIGNALPROCESSINGONGRAPHSCHAPTER8GraphSignalProcessing.CHAPTER9SamplingandRecoveryofGraphSignalsCHAPTER10BayesianActivelearningonGraphs.CHAPTER11DesignofGraphFiltersandFilterbanksCHAPTER12StatisticalGraphSignalProcessing:StationarityandSpectralEstimationCHAPTER13InferenceofGraphTopologyCHAPTER14PartiallyAbsorbingRanclomWalks:AUnifieclFrameworkforLearningonGraphsPART3DISTRIBUTEDCOMMUNICATIONS,NETWORKING,ANDSENSING.....

optimizationmodelsbycalafiore

MIMOOFDMSimulator:OFDM.m:OFDMSimulator(outerfunction)create_channel.m:GeneratesaRayleighfadingfrequency-selectivechannel,parametrizedbytheantennaconfiguration,theOFDMconfiguration,andthepower-delayprofile.svd_decompose_channel.m:Sincefullchannelknowledgeisassumed,transmissionisacrossparallelsingularvaluemodes.Thisfunctiondecomposesthechannelintothesemodes.BitLoad.m:Applythebit-loadingalgorithmtoachievethedesiredbitandenergyallocationforthecurrentchannelinstance.ComputeSNR.m:Giventhesubcarriergains,thissimplefunctiongeneratestheSNRvaluesofeachchannel(eachsingularvalueoneachtoneisaseparatechannel).chow_algo.m:ApplyChow'salgorithmtogenerateaparticularbitandenergyallocation.EnergyTableInit.m:GiventheSNRvalues,formatableofenergyincrementsforeachchannel.campello_algo.m:ApplyCampello'salgorithmtoconvergetotheoptimalbitandenergyallocationforthegivenchannelconditions.ResolvetheLastBit.m:Anoptimalbit-loadingofthelastbitrequiresauniqueoptimization.modulate.m:Modulatetherandominputsequenceaccordingtothebitallocationsforeachchannel.ENC2.mat:BPSKModulatorENC4.mat:4-QAMModulator(Graycoded)ENC16.mat:16-QAMModulator(Graycoded)ENC64.mat:64-QAMModulator(Graycoded)ENC256.mat:256-QAMModulator(Graycoded)precode.m:Precodethetransmittedvectorateachtimeinstancebyfilteringthemodulatedvectorwiththeright-inverseofthechannel'srightsingluarmatrix.ifft_cp_tx_blk.m:IFFTblockoftheOFDMsystem.channel.m:ApplythechanneltotheOFDMframe.fft_cp_rx_blk.m:FFTblockoftheOFDMsystem.shape.m:Completethediagonalizationofthechannelbyfilteringthereceivedvectorwiththeleft-inverseofthechannel'sleftsingularmatrix.demodulate.m:Performanearestneighborsearchknowingthetransmitconstellationused.

FPGA上运行人脸识别源代码，ThisprojectattemptstorealizeafacedetectorusingVoila-Jonesalgorithm.ThereferenceCmodelisborrowedfrom[5kk73GPUAssignment2012](https://sites.google.com/site/5kk73gpu2012/assignment/viola-jones-face-detection),withsomemodifytofithardwareimplementationandfixedsomebug.ThecodeiswrittenbyVerilog/SystemVerilogandSynthesizedonXilinxKintexUltrascaleFPGAusingVivado.Thiscodeisjustexperimentalforfunction,alotofoptimizationcanfurtherbedone.

Thispaperisasurveyofthetheoryandmethodsofphotogrammetricbundleadjustment,aimedatpotentialimplementorsinthecomputervisioncommunity.Bundleadjustmentistheproblemofrefiningavisualreconstructiontoproducejointlyoptimalstructureandviewingparameterestimates.Topicscoveredinclude:thechoiceofcostfunctionandrobustness;numericaloptimizationincludingsparseNewtonmethods,linearlyconvergentapproximations,updatingandrecursivemethods;gauge(datum)invariance;andqualitycontrol.Thetheoryisdevelopedforgeneralrobustcostfunctionsratherthanrestrictingattentiontotraditionalnonlinearleastsquares.

在日常工作中，钉钉打卡成了我生活中不可或缺的一部分。然而，有时候这个看似简单的任务却给我带来了不少烦恼。 每天早晚，我总是得牢记打开钉钉应用，点击"工作台"，再找到"考勤打卡"进行签到。有时候因为工作忙碌，会忘记打卡，导致考勤异常，影响当月的工作评价。而且，由于我使用的是苹果手机，有时候系统更新后，钉钉的某些功能会出现异常，使得打卡变得更加麻烦。 另外，我的家人使用的是安卓手机，他们也经常抱怨钉钉打卡的繁琐。尤其是对于那些不太熟悉手机操作的长辈来说，每次打卡都是一次挑战。他们总是担心自己会操作失误，导致打卡失败。 为了解决这些烦恼，我开始思考是否可以通过编写一个全自动化脚本来实现钉钉打卡。经过一段时间的摸索和学习，我终于成功编写出了一个适用于苹果和安卓系统的钉钉打卡脚本。