Introducing Machine Learning

Dino Esposito: If I look back, I count more 20 books authored and 1000+ articles in a 25-year-long career. I’ve been writing the “Cutting Edge” column for MSDN Magazine month after month for 22 consecutive years. It is commonly recognized that such books and articles have helped the professional growth of thousands of .NET and ASP.NET developers and software architects worldwide.After I escaped a dreadful COBOL project, in 1992 I started as a C developer, and since then, I have witnessed MFC and ATL, COM and DCOM, the debut of .NET, the rise and fall of Silverlight, and the ups and downs of various architectural patterns. In 1995 I led a team of five dreamers who actually deployed things that today we would call Google Photos and Shuttershock–desktop applications capable of dealing with photos stored in a virtual place that nobody had called the cloud yet. Since 2003 I have written Microsoft Press books about ASP.NET and also authored the bestseller Microsoft .NET: Architecting Applications for the Enterprise. I have a few successful Pluralsight courses on .NET architecture, ASP.NET MVC UI, and, recently, ML.NET. As architect of most of the backoffice applications that keep the professional tennis world tour running, I’ve been focusing on renewable energy, IoT, and artificial intelligence for the past two years as the corporate digital strategist at BaxEnergy.You can get in touch with me through https://youbiquitous.net or twitter.com/despos, or you can connect to my LinkedIn network.Francesco Esposito: I was 12 or so in the early days of the Windows Phone launch, and I absolutely wanted one of those devices in my hands. I could have asked Dad or Mom to buy it, but I didn’t know how they would react. As a normal teenager, I had exactly zero chance of having someone buy it for me. So, I found out I was quite good at making sense of programming languages and impressed some folks at Microsoft enough to have a device to test. A Windows Phone was only the beginning; then came my insane passion for iOS and, later, the shortcuts of C#.The current part of my life began when I graduated from high school, one year earlier than expected. By the way, only 0.006 percent of students do that in Italy. I felt as powerful as a semi-god and enrolled in mathematics. I failed my first exams, and the shock put me at work day and night on ASP.NET as a self punishment. I founded my small software company, Youbiquitous, and began living on my own money. In 2017, my innate love for mathematics was resurrected and put me back on track with studies and led me to take the plunge in financial investments and machine learning.This book, then, is the natural consequence of the end of my childhood. I wanted to give something back to my dad and help him make sense of the deep mathematics behind neural networks and algorithms. By the way, I have a dream: developing a supertheory of intelligence that would mathematically explore why the artificial intelligence of today works and where we can go further.You can get in touch with me at https://youbiquitous.net.

• IntroductionPart I Laying the Groundwork of Machine LearningChapter 1 How Humans LearnThe Journey Toward Thinking MachinesThe Dawn of Mechanical ReasoningGodel’s Incompleteness TheoremsFormalization of Computing MachinesToward the Formalization of Human ThoughtThe Birth of Artificial Intelligence as a DisciplineThe Biology of LearningWhat Is Intelligent Software, Anyway?How Neurons WorkThe Carrot-and-Stick ApproachAdaptability to ChangesArtificial Forms of IntelligencePrimordial IntelligenceExpert SystemsAutonomous SystemsArtificial Forms of SentimentSummaryChapter 2 Intelligent SoftwareApplied Artificial IntelligenceEvolution of Software IntelligenceExpert SystemsGeneral Artificial IntelligenceUnsupervised LearningSupervised LearningSummaryChapter 3 Mapping Problems and AlgorithmsFundamental ProblemsClassifying ObjectsPredicting ResultsGrouping ObjectsMore Complex ProblemsImage ClassificationObject DetectionText AnalyticsAutomated Machine LearningAspects of an AutoML PlatformThe AutoML Model Builder in ActionSummaryChapter 4 General Steps for a Machine Learning SolutionData CollectionData-Driven Culture in the OrganizationStorage OptionsData PreparationImproving Data QualityCleaning DataFeature EngineeringFinalizing the Training DatasetModel Selection and TrainingThe Algorithm Cheat SheetThe Case for Neural NetworksEvaluation of the Model PerformanceDeployment of the ModelChoosing the Appropriate Hosting PlatformExposing an APISummaryChapter 5 The Data FactorData QualityData ValidityData CollectionData IntegrityCompletenessUniquenessTimelinessAccuracyConsistencyWhat’s a Data Scientist, Anyway?The Data Scientist at WorkThe Data Scientist Tool ChestData Scientists and Software DevelopersSummaryPart II Machine Learning In .NETChapter 6 The .NET WayWhy (Not) Python?Why Is Python So Popular in Machine Learning?Taxonomy of Python Machine Learning LibrariesEnd-to-End Solutions on Top of Python ModelsIntroducing ML.NETCreating and Consuming Models in ML.NETElements of the Learning ContextSummaryChapter 7 Implementing the ML.NET PipelineThe Data to Start FromExploring the DatasetApplying Common Data TransformationsConsiderations on the DatasetThe Training StepPicking an AlgorithmMeasuring the Actual Value of an AlgorithmPlanning the Testing PhaseA Look at the MetricsPrice Prediction from Within a Client ApplicationGetting the Model FileSetting Up the ASP.NET ApplicationMaking a Taxi Fare PredictionDevising an Adequate User InterfaceQuestioning Data and Approach to the ProblemSummaryChapter 8 ML.NET Tasks and AlgorithmsThe Overall ML.NET ArchitectureInvolved Types and InterfacesData RepresentationSupported CatalogsClassification TasksBinary ClassificationMulticlass ClassificationClustering TasksPreparing Data for WorkTraining the ModelEvaluating the ModelTransfer LearningSteps for Building an Image ClassifierApplying Necessary Data TransformationsComposing and Training the ModelMargin Notes on Transfer LearningSummaryPart III Fundamentals of Shallow LearningChapter 9 Math Foundations of Machine LearningUnder the Umbrella of StatisticsThe Mean in StatisticsThe Mode in StatisticsThe Median in StatisticsBias and VarianceThe Variance in StatisticsThe Bias in StatisticsData RepresentationFive-number SummaryHistogramsScatter PlotsScatter Plot MatricesPlotting at the Appropriate ScaleSummaryChapter 10 Metrics of Machine LearningStatistics vs. Machine LearningThe Ultimate Goal of Machine LearningFrom Statistical Models to Machine Learning ModelsEvaluation of a Machine Learning ModelFrom Dataset to PredictionsMeasuring the Precision of a ModelPreparing Data for ProcessingScalingStandardizationNormalizationSummaryChapter 11 How to Make Simple Predictions: Linear RegressionThe ProblemGuessing Results Guided by DataMaking Hypotheses About the RelationshipThe Linear AlgorithmThe General IdeaIdentifying the Cost FunctionThe Ordinary Least Square AlgorithmThe Gradient Descent AlgorithmHow Good Is the Algorithm?Improving the SolutionThe Polynomial RouteRegularizationSummaryChapter 12 How to Make Complex Predictions and Decisions: TreesThe ProblemWhat’s a Tree, Anyway?Trees in Machine LearningA Sample Tree-Based AlgorithmDesign Principles for Tree-Based AlgorithmsDecision Trees versus Expert SystemsFlavors of Tree AlgorithmsClassification TreesHow the CART Algorithm WorksHow the ID3 Algorithm WorksRegression TreesHow the Algorithm WorksTree PruningSummaryChapter 13 How to Make Better Decisions: Ensemble MethodsThe ProblemThe Bagging TechniqueRandom Forest AlgorithmsSteps of the AlgorithmsPros and ConsThe Boosting TechniqueThe Power of BoostingGradient BoostingPros and ConsSummaryChapter 14 Probabilistic Methods: Naïve BayesQuick Introduction to Bayesian StatisticsIntroducing Bayesian ProbabilitySome Preliminary NotationBayes’ TheoremA Practical Code Review ExampleApplying Bayesian Statistics to ClassificationInitial Formulation of the ProblemA Simplified (Yet Effective) FormulationPractical Aspects of Bayesian ClassifiersNaïve Bayes ClassifiersThe General AlgorithmMultinomial Naïve BayesBernoulli Naïve BayesGaussian Naïve BayesNaïve Bayes RegressionFoundation of Bayesian Linear RegressionApplications of Bayesian Linear RegressionSummaryChapter 15 How to Group Data: Classification and ClusteringA Basic Approach to Supervised ClassificationThe K-Nearest Neighbors AlgorithmSteps of the AlgorithmBusiness ScenariosSupport Vector MachineOverview of the AlgorithmA Quick Mathematical RefresherSteps of the AlgorithmUnsupervised ClusteringA Business Case: Reducing the DatasetThe K-Means AlgorithmThe K-Modes AlgorithmThe DBSCAN AlgorithmSummaryPart IV Fundamentals of Deep LearningChapter 16 Feed-Forward Neural NetworksA Brief History of Neural NetworksThe McCulloch-Pitt NeuronFeed-Forward NetworksMore Sophisticated NetworksTypes of Artificial NeuronsThe Perceptron NeuronThe Logistic NeuronTraining a Neural NetworkThe Overall Learning StrategyThe Backpropagation AlgorithmSummaryChapter 17 Design of a Neural NetworkAspects of a Neural NetworkActivation FunctionsHidden LayersThe Output LayerBuilding a Neural NetworkAvailable FrameworksYour First Neural Network in KerasNeural Networks versus Other AlgorithmsSummaryChapter 18 Other Types of Neural NetworksCommon Issues of Feed-Forward Neural NetworksRecurrent Neural NetworksAnatomy of a Stateful Neural NetworkLSTM Neural NetworksConvolutional Neural NetworksImage Classification and RecognitionThe Convolutional LayerThe Pooling LayerThe Fully Connected LayerFurther Neural Network DevelopmentsGenerative Adversarial Neural NetworksAuto-EncodersSummaryChapter 19 Sentiment Analysis: An End-to-End SolutionPreparing Data for TrainingFormalizing the ProblemGetting the Data.Manipulating the DataConsiderations on the Intermediate FormatTraining the ModelChoosing the EcosystemBuilding a Dictionary of WordsChoosing the TrainerOther Aspects of the NetworkThe Client ApplicationGetting Input for the ModelGetting the Prediction from the ModelTurning the Response into Usable InformationSummaryPart V Final ThoughtsChapter 20 AI Cloud Services for the Real WorldAzure Cognitive ServicesAzure Machine Learning StudioAzure Machine Learning ServiceData Science Virtual MachinesOn-Premises ServicesSQL Server Machine Learning ServicesMachine Learning ServerMicrosoft Data Processing ServicesAzure Data LakeAzure DatabricksAzure HDInsight.NET for Apache SparkAzure Data ShareAzure Data FactorySummaryChapter 21 The Business Perception of AIPerception of AI in the IndustryRealizing the PotentialWhat Artificial Intelligence Can Do for YouChallenges Around the CornerEnd-to-End SolutionsLet’s Just Call It ConsultingThe Borderline Between Software and Data ScienceAgile AISummary9780135565667 TOC 12/19/2019