{"product_id":"fundamentals-of-robust-machine-learning-isbn-9781394294374","title":"Fundamentals of Robust Machine Learning","description":"\u003cp\u003e\u003cb\u003eAn essential guide for tackling outliers and anomalies in machine learning and data science.\u003c\/b\u003e \u003c\/p\u003e\u003cp\u003eIn recent years, machine learning (ML) has transformed virtually every area of research and technology, becoming one of the key tools for data scientists. Robust machine learning is a new approach to handling outliers in datasets, which is an often-overlooked aspect of data science. Ignoring outliers can lead to bad business decisions, wrong medical diagnoses, reaching the wrong conclusions or incorrectly assessing feature importance, just to name a few. \u003c\/p\u003e\u003cp\u003e\u003ci\u003eFundamentals of Robust Machine Learning\u003c\/i\u003e offers a thorough but accessible overview of this subject by focusing on how to properly handle outliers and anomalies in datasets. There are two main approaches described in the book: using outlier-tolerant ML tools, or removing outliers before using conventional tools. Balancing theoretical foundations with practical Python code, it provides all the necessary skills to enhance the accuracy, stability and reliability of ML models. \u003c\/p\u003e\u003cp\u003e\u003ci\u003eFundamentals of Robust Machine Learning \u003c\/i\u003ereaders will also find: \u003c\/p\u003e\u003cul\u003e\n\u003cli\u003eA blend of robust statistics and machine learning principles\u003c\/li\u003e\n\u003cli\u003eDetailed discussion of a wide range of robust machine learning methodologies, from robust clustering, regression and classification, to neural networks and anomaly detection\u003c\/li\u003e\n\u003cli\u003ePython code with immediate application to data science problems\u003c\/li\u003e\n\u003c\/ul\u003e \u003cp\u003e\u003ci\u003eFundamentals of Robust Machine Learning\u003c\/i\u003e is ideal for undergraduate or graduate students in data science, machine learning, and related fields, as well as for professionals in the field looking to enhance their understanding of building models in the presence of outliers. \u003c\/p\u003e\u003cp\u003ePreface xv\u003c\/p\u003e \u003cp\u003eAbout the Companion Website xix\u003c\/p\u003e \u003cp\u003e\u003cb\u003e1 Introduction 1\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e1.1 Defining Outliers 2\u003c\/p\u003e \u003cp\u003e1.2 Overview of the Book 3\u003c\/p\u003e \u003cp\u003e1.3 What Is Robust Machine Learning? 3\u003c\/p\u003e \u003cp\u003e1.3.1 Machine Learning Basics 4\u003c\/p\u003e \u003cp\u003e1.3.2 Effect of Outliers 6\u003c\/p\u003e \u003cp\u003e1.3.3 What Is Robust Data Science? 7\u003c\/p\u003e \u003cp\u003e1.3.4 Noise in Datasets 7\u003c\/p\u003e \u003cp\u003e1.3.5 Training and Testing Flows 8\u003c\/p\u003e \u003cp\u003e1.4 Robustness of the Median 9\u003c\/p\u003e \u003cp\u003e1.4.1 Mean vs. Median 9\u003c\/p\u003e \u003cp\u003e1.4.2 Effect on Standard Deviation 10\u003c\/p\u003e \u003cp\u003e1.5 l 1 and l 2 Norms 11\u003c\/p\u003e \u003cp\u003e1.6 Review of Gaussian Distribution 12\u003c\/p\u003e \u003cp\u003e1.7 Unsupervised Learning Case Study 13\u003c\/p\u003e \u003cp\u003e1.7.1 Clustering Example 14\u003c\/p\u003e \u003cp\u003e1.7.2 Clustering Problem Specification 14\u003c\/p\u003e \u003cp\u003e1.8 Creating Synthetic Data for Clustering 16\u003c\/p\u003e \u003cp\u003e1.8.1 One-Dimensional Datasets 16\u003c\/p\u003e \u003cp\u003e1.8.2 Multidimensional Datasets 17\u003c\/p\u003e \u003cp\u003e1.9 Clustering Algorithms 19\u003c\/p\u003e \u003cp\u003e1.9.1 k-Means Clustering 19\u003c\/p\u003e \u003cp\u003e1.9.2 k-Medians Clustering 21\u003c\/p\u003e \u003cp\u003e1.10 Importance of Robust Clustering 22\u003c\/p\u003e \u003cp\u003e1.10.1 Clustering with No Outliers 22\u003c\/p\u003e \u003cp\u003e1.10.2 Clustering with Outliers 23\u003c\/p\u003e \u003cp\u003e1.10.3 Detection and Removal of Outliers 25\u003c\/p\u003e \u003cp\u003e1.11 Summary 27\u003c\/p\u003e \u003cp\u003eProblems 28\u003c\/p\u003e \u003cp\u003eReferences 34\u003c\/p\u003e \u003cp\u003e\u003cb\u003e2 Robust Linear Regression 35\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e2.1 Introduction 35\u003c\/p\u003e \u003cp\u003e2.2 Supervised Learning 35\u003c\/p\u003e \u003cp\u003e2.3 Linear Regression 36\u003c\/p\u003e \u003cp\u003e2.4 Importance of Residuals 38\u003c\/p\u003e \u003cp\u003e2.4.1 Defining Errors and Residuals 38\u003c\/p\u003e \u003cp\u003e2.4.2 Residuals in Loss Functions 39\u003c\/p\u003e \u003cp\u003e2.4.3 Distribution of Residuals 40\u003c\/p\u003e \u003cp\u003e2.5 Estimation Background 42\u003c\/p\u003e \u003cp\u003e2.5.1 Linear Models 42\u003c\/p\u003e \u003cp\u003e2.5.2 Desirable Properties of Estimators 43\u003c\/p\u003e \u003cp\u003e2.5.3 Maximum-Likelihood Estimation 44\u003c\/p\u003e \u003cp\u003e2.5.4 Gradient Descent 47\u003c\/p\u003e \u003cp\u003e2.6 M-Estimation 49\u003c\/p\u003e \u003cp\u003e2.7 Least Squares Estimation (LSE) 52\u003c\/p\u003e \u003cp\u003e2.8 Least Absolute Deviation (LAD) 54\u003c\/p\u003e \u003cp\u003e2.9 Comparison of LSE and LAD 55\u003c\/p\u003e \u003cp\u003e2.9.1 Simple Linear Model 55\u003c\/p\u003e \u003cp\u003e2.9.2 Location Problem 56\u003c\/p\u003e \u003cp\u003e2.10 Huber’s Method 58\u003c\/p\u003e \u003cp\u003e2.10.1 Huber Loss Function 58\u003c\/p\u003e \u003cp\u003e2.10.2 Comparison with LSE and LAD 63\u003c\/p\u003e \u003cp\u003e2.11 Summary 64\u003c\/p\u003e \u003cp\u003eProblems 64\u003c\/p\u003e \u003cp\u003eReferences 67\u003c\/p\u003e \u003cp\u003e\u003cb\u003e3 The Log-Cosh Loss Function 69\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e3.1 Introduction 69\u003c\/p\u003e \u003cp\u003e3.2 An Intuitive View of Log-Cosh 69\u003c\/p\u003e \u003cp\u003e3.3 Hyperbolic Functions 71\u003c\/p\u003e \u003cp\u003e3.4 M-Estimation 71\u003c\/p\u003e \u003cp\u003e3.4.1 Asymptotic Behavior 72\u003c\/p\u003e \u003cp\u003e3.4.2 Linear Regression Using Log-Cosh 74\u003c\/p\u003e \u003cp\u003e3.5 Deriving the Distribution for Log-Cosh 75\u003c\/p\u003e \u003cp\u003e3.6 Standard Errors for Robust Estimators 79\u003c\/p\u003e \u003cp\u003e3.6.1 Example: Swiss Fertility Dataset 81\u003c\/p\u003e \u003cp\u003e3.6.2 Example: Boston Housing Dataset 82\u003c\/p\u003e \u003cp\u003e3.7 Statistical Properties of Log-Cosh Loss 83\u003c\/p\u003e \u003cp\u003e3.7.1 Maximum-Likelihood Estimation 83\u003c\/p\u003e \u003cp\u003e3.8 A General Log-Cosh Loss Function 84\u003c\/p\u003e \u003cp\u003e3.9 Summary 88\u003c\/p\u003e \u003cp\u003eProblems 88\u003c\/p\u003e \u003cp\u003eReferences 93\u003c\/p\u003e \u003cp\u003e\u003cb\u003e4 Outlier Detection, Metrics, and Standardization 95\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e4.1 Introduction 95\u003c\/p\u003e \u003cp\u003e4.2 Effect of Outliers 95\u003c\/p\u003e \u003cp\u003e4.3 Outlier Diagnosis 97\u003c\/p\u003e \u003cp\u003e4.3.1 Boxplots 98\u003c\/p\u003e \u003cp\u003e4.3.2 Histogram Plots 100\u003c\/p\u003e \u003cp\u003e4.3.3 Exploratory Data Analysis 101\u003c\/p\u003e \u003cp\u003e4.4 Outlier Detection 102\u003c\/p\u003e \u003cp\u003e4.4.1 3-Sigma Edit Rule 102\u003c\/p\u003e \u003cp\u003e4.4.2 4.5-MAD Edit Rule 104\u003c\/p\u003e \u003cp\u003e4.4.3 1.5-IQR Edit Rule 105\u003c\/p\u003e \u003cp\u003e4.5 Outlier Removal 105\u003c\/p\u003e \u003cp\u003e4.5.1 Trimming Methods 105\u003c\/p\u003e \u003cp\u003e4.5.2 Winsorization 105\u003c\/p\u003e \u003cp\u003e4.5.3 Anomaly Detection Method 106\u003c\/p\u003e \u003cp\u003e4.6 Regression-Based Outlier Detection 107\u003c\/p\u003e \u003cp\u003e4.6.1 LS vs. LC Residuals 108\u003c\/p\u003e \u003cp\u003e4.6.2 Comparison of Detection Methods 109\u003c\/p\u003e \u003cp\u003e4.6.3 Ordered Absolute Residuals (OARs) 110\u003c\/p\u003e \u003cp\u003e4.6.4 Quantile–Quantile Plot 111\u003c\/p\u003e \u003cp\u003e4.6.5 Quad-Plots for Outlier Diagnosis 113\u003c\/p\u003e \u003cp\u003e4.7 Regression-Based Outlier Removal 114\u003c\/p\u003e \u003cp\u003e4.7.1 Iterative Boxplot Method 114\u003c\/p\u003e \u003cp\u003e4.8 Regression Metrics with Outliers 116\u003c\/p\u003e \u003cp\u003e4.8.1 Mean Square Error (MSE) 117\u003c\/p\u003e \u003cp\u003e4.8.2 Median Absolute Error (MAE) 118\u003c\/p\u003e \u003cp\u003e4.8.3 MSE vs. MAE on Realistic Data 119\u003c\/p\u003e \u003cp\u003e4.8.4 Selecting Hyperparameters for Robust Regression 120\u003c\/p\u003e \u003cp\u003e4.9 Dataset Standardization 121\u003c\/p\u003e \u003cp\u003e4.9.1 Robust Standardization 122\u003c\/p\u003e \u003cp\u003e4.10 Summary 126\u003c\/p\u003e \u003cp\u003eProblems 126\u003c\/p\u003e \u003cp\u003eReferences 131\u003c\/p\u003e \u003cp\u003e\u003cb\u003e5 Robustness of Penalty Estimators 133\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e5.1 Introduction 133\u003c\/p\u003e \u003cp\u003e5.2 Penalty Functions 133\u003c\/p\u003e \u003cp\u003e5.2.1 Multicollinearity 133\u003c\/p\u003e \u003cp\u003e5.2.2 Penalized Loss Functions 135\u003c\/p\u003e \u003cp\u003e5.3 Ridge Penalty 136\u003c\/p\u003e \u003cp\u003e5.4 LASSO Penalty 137\u003c\/p\u003e \u003cp\u003e5.5 Effect of Penalty Functions 138\u003c\/p\u003e \u003cp\u003e5.6 Penalty Functions with Outliers 139\u003c\/p\u003e \u003cp\u003e5.7 Ridge Traces 142\u003c\/p\u003e \u003cp\u003e5.8 Elastic Net (Enet) Penalty 143\u003c\/p\u003e \u003cp\u003e5.9 Adaptive LASSO (aLASSO) Penalty 145\u003c\/p\u003e \u003cp\u003e5.10 Penalty Effects on Variance and Bias 146\u003c\/p\u003e \u003cp\u003e5.10.1 Effect on Variance 146\u003c\/p\u003e \u003cp\u003e5.10.2 Geometric Interpretation of Bias 148\u003c\/p\u003e \u003cp\u003e5.11 Variable Importance 151\u003c\/p\u003e \u003cp\u003e5.11.1 The t-Statistic 151\u003c\/p\u003e \u003cp\u003e5.11.2 LASSO and aLASSO Traces 153\u003c\/p\u003e \u003cp\u003e5.12 Summary 155\u003c\/p\u003e \u003cp\u003eProblems 156\u003c\/p\u003e \u003cp\u003eReferences 159\u003c\/p\u003e \u003cp\u003e\u003cb\u003e6 Robust Regularized Models 161\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e6.1 Introduction 161\u003c\/p\u003e \u003cp\u003e6.2 Overfitting and Underfitting 161\u003c\/p\u003e \u003cp\u003e6.3 The Bias–Variance Trade-Off 162\u003c\/p\u003e \u003cp\u003e6.4 Regularization with Ridge 164\u003c\/p\u003e \u003cp\u003e6.4.1 Selection of Hyperparameter λ 165\u003c\/p\u003e \u003cp\u003e6.4.2 Example: Diabetes Dataset 167\u003c\/p\u003e \u003cp\u003e6.5 Generalization using Robust Estimators 169\u003c\/p\u003e \u003cp\u003e6.5.1 Training and Test Sets 169\u003c\/p\u003e \u003cp\u003e6.5.2 k-Fold Cross-validation 171\u003c\/p\u003e \u003cp\u003e6.6 Robust Generalization and Regularization 173\u003c\/p\u003e \u003cp\u003e6.6.1 Regularization with LC-Ridge 174\u003c\/p\u003e \u003cp\u003e6.7 Model Complexity 175\u003c\/p\u003e \u003cp\u003e6.7.1 Variable Selection Using LS-LASSO 176\u003c\/p\u003e \u003cp\u003e6.7.2 Variable Ordering Using LC-aLASSO 176\u003c\/p\u003e \u003cp\u003e6.7.3 Building a Compact Model 179\u003c\/p\u003e \u003cp\u003e6.8 Summary 182\u003c\/p\u003e \u003cp\u003eProblems 182\u003c\/p\u003e \u003cp\u003eReferences 186\u003c\/p\u003e \u003cp\u003e\u003cb\u003e7 Quantile Regression Using Log-Cosh 187\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e7.1 Introduction 187\u003c\/p\u003e \u003cp\u003e7.2 Understanding Quantile Regression 188\u003c\/p\u003e \u003cp\u003e7.3 The Crossing Problem 189\u003c\/p\u003e \u003cp\u003e7.4 Standard Quantile Loss Function 190\u003c\/p\u003e \u003cp\u003e7.5 Smooth Regression Quantiles (SMRQ) 192\u003c\/p\u003e \u003cp\u003e7.6 Evaluation of Quantile Methods 195\u003c\/p\u003e \u003cp\u003e7.6.1 Qualitative Assessment 196\u003c\/p\u003e \u003cp\u003e7.6.2 Quantitative Assessment 198\u003c\/p\u003e \u003cp\u003e7.7 Selection of Robustness Coefficient 200\u003c\/p\u003e \u003cp\u003e7.8 Maximum-Likelihood Procedure for SMRQ 202\u003c\/p\u003e \u003cp\u003e7.9 Standard Error Computation 204\u003c\/p\u003e \u003cp\u003e7.10 Summary 206\u003c\/p\u003e \u003cp\u003eProblems 207\u003c\/p\u003e \u003cp\u003eReferences 209\u003c\/p\u003e \u003cp\u003e\u003cb\u003e8 Robust Binary Classification 211\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e8.1 Introduction 211\u003c\/p\u003e \u003cp\u003e8.2 Binary Classification Problem 212\u003c\/p\u003e \u003cp\u003e8.2.1 Why Linear Regression Fails 212\u003c\/p\u003e \u003cp\u003e8.2.2 Outliers in Binary Classification 213\u003c\/p\u003e \u003cp\u003e8.3 The Cross-Entropy (CE) Loss 215\u003c\/p\u003e \u003cp\u003e8.3.1 Deriving the Cross-Entropy Loss 216\u003c\/p\u003e \u003cp\u003e8.3.2 Understanding Logistic Regression 218\u003c\/p\u003e \u003cp\u003e8.3.3 Gradient Descent 221\u003c\/p\u003e \u003cp\u003e8.4 The Log-Cosh (LC) Loss Function 221\u003c\/p\u003e \u003cp\u003e8.4.1 General Formulation 223\u003c\/p\u003e \u003cp\u003e8.5 Algorithms for Logistic Regression 224\u003c\/p\u003e \u003cp\u003e8.6 Example: Motor Trend Cars 226\u003c\/p\u003e \u003cp\u003e8.7 Regularization of Logistic Regression 227\u003c\/p\u003e \u003cp\u003e8.7.1 Overfitting and Underfitting 228\u003c\/p\u003e \u003cp\u003e8.7.2 k-Fold Cross-Validation 229\u003c\/p\u003e \u003cp\u003e8.7.3 Penalty Functions 229\u003c\/p\u003e \u003cp\u003e8.7.4 Effect of Outliers 230\u003c\/p\u003e \u003cp\u003e8.8 Example: Circular Dataset 231\u003c\/p\u003e \u003cp\u003e8.9 Outlier Detection 234\u003c\/p\u003e \u003cp\u003e8.10 Robustness of Binary Classifiers 235\u003c\/p\u003e \u003cp\u003e8.10.1 Support Vector Classifier (SVC) 235\u003c\/p\u003e \u003cp\u003e8.10.2 Support Vector Machines (SVMs) 238\u003c\/p\u003e \u003cp\u003e8.10.3 k-Nearest Neighbors (k-NN) 241\u003c\/p\u003e \u003cp\u003e8.10.4 Decision Trees and Random Forest 243\u003c\/p\u003e \u003cp\u003e8.11 Summary 244\u003c\/p\u003e \u003cp\u003eProblems 244\u003c\/p\u003e \u003cp\u003eReference 249\u003c\/p\u003e \u003cp\u003e\u003cb\u003e9 Neural Networks Using Log-Cosh 251\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e9.1 Introduction 251\u003c\/p\u003e \u003cp\u003e9.2 A Brief History of Neural Networks 251\u003c\/p\u003e \u003cp\u003e9.3 Defining Neural Networks 252\u003c\/p\u003e \u003cp\u003e9.3.1 Basic Computational Unit 253\u003c\/p\u003e \u003cp\u003e9.3.2 Four-Layer Neural Network 254\u003c\/p\u003e \u003cp\u003e9.3.3 Activation Functions 255\u003c\/p\u003e \u003cp\u003e9.4 Training of Neural Networks 257\u003c\/p\u003e \u003cp\u003e9.5 Forward and Backward Propagation 258\u003c\/p\u003e \u003cp\u003e9.5.1 Forward Propagation 259\u003c\/p\u003e \u003cp\u003e9.5.2 Backward Propagation 260\u003c\/p\u003e \u003cp\u003e9.5.3 Log-Cosh Gradients 263\u003c\/p\u003e \u003cp\u003e9.6 Cross-entropy and Log-Cosh Algorithms 264\u003c\/p\u003e \u003cp\u003e9.7 Example: Circular Dataset 266\u003c\/p\u003e \u003cp\u003e9.8 Classification Metrics and Outliers 269\u003c\/p\u003e \u003cp\u003e9.8.1 Precision, Recall, F 1 Score 269\u003c\/p\u003e \u003cp\u003e9.8.2 Receiver Operating Characteristics (ROCs) 271\u003c\/p\u003e \u003cp\u003e9.9 Summary 273\u003c\/p\u003e \u003cp\u003eProblems 273\u003c\/p\u003e \u003cp\u003eReferences 280\u003c\/p\u003e \u003cp\u003e\u003cb\u003e10 Multi-class Classification and Adam Optimization 281\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e10.1 Introduction 281\u003c\/p\u003e \u003cp\u003e10.2 Multi-class Classification 281\u003c\/p\u003e \u003cp\u003e10.2.1 Multi-class Loss Functions 282\u003c\/p\u003e \u003cp\u003e10.2.2 Softmax Activation Function 284\u003c\/p\u003e \u003cp\u003e10.3 Example: MNIST Dataset 288\u003c\/p\u003e \u003cp\u003e10.3.1 Neural Network Architecture 289\u003c\/p\u003e \u003cp\u003e10.3.2 Comparing Cross-Entropy with Log-Cosh Losses 289\u003c\/p\u003e \u003cp\u003e10.3.3 Outliers in MNIST 291\u003c\/p\u003e \u003cp\u003e10.4 Optimization of Neural Networks 291\u003c\/p\u003e \u003cp\u003e10.4.1 Momentum 293\u003c\/p\u003e \u003cp\u003e10.4.2 rmsprop Approach 294\u003c\/p\u003e \u003cp\u003e10.4.3 Optimizer Warm-Up Phase 295\u003c\/p\u003e \u003cp\u003e10.4.4 Adam Optimizer 296\u003c\/p\u003e \u003cp\u003e10.5 Summary 297\u003c\/p\u003e \u003cp\u003eProblems 297\u003c\/p\u003e \u003cp\u003eReferences 302\u003c\/p\u003e \u003cp\u003e\u003cb\u003e11 Anomaly Detection and Evaluation Metrics 303\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e11.1 Introduction 303\u003c\/p\u003e \u003cp\u003e11.2 Anomaly Detection Methods 303\u003c\/p\u003e \u003cp\u003e11.2.1 k-Nearest Neighbors 304\u003c\/p\u003e \u003cp\u003e11.2.2 Dbscan 308\u003c\/p\u003e \u003cp\u003e11.2.3 Isolation Forest 311\u003c\/p\u003e \u003cp\u003e11.3 Anomaly Detection Using MADmax 316\u003c\/p\u003e \u003cp\u003e11.3.1 Robust Standardization 317\u003c\/p\u003e \u003cp\u003e11.3.2 k-Medians Clustering 317\u003c\/p\u003e \u003cp\u003e11.3.3 Selecting MADmax 319\u003c\/p\u003e \u003cp\u003e11.3.4 k-Nearest Neighbors (k-NN) 319\u003c\/p\u003e \u003cp\u003e11.3.5 k-Nearest Medians (k-NM) 320\u003c\/p\u003e \u003cp\u003e11.4 Qualitative Evaluation Methods 323\u003c\/p\u003e \u003cp\u003e11.5 Quantitative Evaluation Methods 326\u003c\/p\u003e \u003cp\u003e11.6 Summary 330\u003c\/p\u003e \u003cp\u003eProblems 330\u003c\/p\u003e \u003cp\u003eReference 336\u003c\/p\u003e \u003cp\u003e\u003cb\u003e12 Case Studies in Data Science 337\u003c\/b\u003e\u003c\/p\u003e \u003cp\u003e12.1 Introduction 337\u003c\/p\u003e \u003cp\u003e12.2 Example: Boston Housing Dataset 337\u003c\/p\u003e \u003cp\u003e12.2.1 Exploratory Data Analysis 338\u003c\/p\u003e \u003cp\u003e12.2.2 Neural Network Architecture 339\u003c\/p\u003e \u003cp\u003e12.2.3 Comparison of LSNN and LCNN 342\u003c\/p\u003e \u003cp\u003e12.2.4 Predicting Housing Prices 344\u003c\/p\u003e \u003cp\u003e12.2.5 RMSE vs. MAE 344\u003c\/p\u003e \u003cp\u003e12.2.6 Correlation Coefficients 345\u003c\/p\u003e \u003cp\u003e12.3 Example: Titanic Dataset 346\u003c\/p\u003e \u003cp\u003e12.3.1 Exploratory Data Analysis 346\u003c\/p\u003e \u003cp\u003e12.3.2 LCLR vs. CELR 351\u003c\/p\u003e \u003cp\u003e12.3.3 Outlier Detection and Removal 353\u003c\/p\u003e \u003cp\u003e12.3.4 Robustness Coefficient for Log-Cosh 355\u003c\/p\u003e \u003cp\u003e12.3.5 The Implications of Robustness 356\u003c\/p\u003e \u003cp\u003e12.3.6 Ridge and aLASSO 357\u003c\/p\u003e \u003cp\u003e12.4 Application to Explainable Artificial Intelligence (XAI) 359\u003c\/p\u003e \u003cp\u003e12.4.1 Case Study: Logistic Regression 360\u003c\/p\u003e \u003cp\u003e12.4.2 Case Study: Neural Networks 365\u003c\/p\u003e \u003cp\u003e12.5 Time Series Example: Climate Change 366\u003c\/p\u003e \u003cp\u003e12.5.1 Autoregressive Model 367\u003c\/p\u003e \u003cp\u003e12.5.2 Forecasting Using AR(p) 369\u003c\/p\u003e \u003cp\u003e12.5.3 Stationary Time Series 371\u003c\/p\u003e \u003cp\u003e12.5.4 Moving Average 374\u003c\/p\u003e \u003cp\u003e12.5.5 Finding Outliers in Time Series 375\u003c\/p\u003e \u003cp\u003e12.6 Summary and Conclusions 376\u003c\/p\u003e \u003cp\u003eProblems 376\u003c\/p\u003e \u003cp\u003eReferences 382\u003c\/p\u003e \u003cp\u003eIndex 383\u003c\/p\u003e  \u003cp\u003e\u003cb\u003eResve Saleh, (PhD, UC Berkeley)\u003c\/b\u003e is a Professor Emeritus at the University of British Columbia. He worked for a decade as a professor at the University of Illinois and as a visiting professor at Stanford University. He was Founder and Chairman of Simplex Solutions, Inc., which went public in 2001. He is an IEEE Fellow and Fellow of the Canadian Academy of Engineering. \u003c\/p\u003e\u003cp\u003e\u003cb\u003eSohaib Majzoub, (PhD, University of British Columbia)\u003c\/b\u003e is an Associate Professor at the University of Sharjah, UAE. He also taught at the American University in Dubai, UAE and at King Saud University, KSA, and a visiting professor at Delft Technical University in The Netherlands. He is a Senior Member of the IEEE. \u003c\/p\u003e\u003cp\u003e\u003cb\u003eA. K. MD. Ehsanes Saleh, (PhD, University of Western Ontario)\u003c\/b\u003e is a Professor Emeritus and Distinguished Professor in the School of Mathematics and Statistics, Carleton University, Ottawa, Canada. He also taught as Simon Fraser University, the University of Toronto, and Stanford University. He is a Fellow of IMS, ASA and an Honorary Member of SSC, Canada.   \u003c\/p\u003e\u003cp\u003e\u003cb\u003eAn essential guide for tackling outliers and anomalies in machine learning and data science.\u003c\/b\u003e \u003c\/p\u003e\u003cp\u003eIn recent years, machine learning (ML) has transformed virtually every area of research and technology, becoming one of the key tools for data scientists. Robust machine learning is a new approach to handling outliers in datasets, which is an often-overlooked aspect of data science. Ignoring outliers can lead to bad business decisions, wrong medical diagnoses, reaching the wrong conclusions or incorrectly assessing feature importance, just to name a few. \u003c\/p\u003e\u003cp\u003e\u003ci\u003eFundamentals of Robust Machine Learning\u003c\/i\u003e offers a thorough but accessible overview of this subject by focusing on how to properly handle outliers and anomalies in datasets. There are two main approaches described in the book: using outlier-tolerant ML tools, or removing outliers before using conventional tools. Balancing theoretical foundations with practical Python code, it provides all the necessary skills to enhance the accuracy, stability and reliability of ML models. \u003c\/p\u003e\u003cp\u003e\u003ci\u003eFundamentals of Robust Machine Learning \u003c\/i\u003ereaders will also find: \u003c\/p\u003e\u003cul\u003e\n\u003cli\u003eA blend of robust statistics and machine learning principles\u003c\/li\u003e\n\u003cli\u003eDetailed discussion of a wide range of robust machine learning methodologies, from robust clustering, regression and classification, to neural networks and anomaly detection\u003c\/li\u003e\n\u003cli\u003ePython code with immediate application to data science problems\u003c\/li\u003e\n\u003c\/ul\u003e \u003cp\u003e\u003ci\u003eFundamentals of Robust Machine Learning\u003c\/i\u003e is ideal for undergraduate or graduate students in data science, machine learning, and related fields, as well as for professionals in the field looking to enhance their understanding of building models in the presence of outliers.\u003c\/p\u003e","brand":"Wiley","offers":[{"title":"Default Title","offer_id":47989263761637,"sku":"NP9781394294374","price":110.0,"currency_code":"USD","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/1842\/7735\/files\/9781394294374.jpg?v=1761783432","url":"https:\/\/k12savings.com\/es\/products\/fundamentals-of-robust-machine-learning-isbn-9781394294374","provider":"K12savings","version":"1.0","type":"link"}