Table of Contents

Preface		xi
Chapter 1	Introduction	1
	1.1 Machine learning vs. statistics	4
	1.2 This book and the ancillary material	5
	1.3 Types of machine learning models	6
	1.4 Validation and testing	8
	1.5 Data cleaning	16
	1.6 Bayes’ theorem	18
	Summary	20
	Short concept questions	21
	Exercises	22
Chapter 2	Unsupervised Learning	23
	2.1 Feature scaling	24
	2.2 The k-means algorithm	25
	2.3 Choosing k	30
	2.4 The curse of dimensionality	33
	2.5 Country risk	34
	2.6 Alternative clustering approaches	39
	2.7 Principal components analysis	41
	Summary	45
	Short concept questions	46
	Exercises	47
Chapter 3	Supervised Learning: Linear and Logistic Regression	49
	3.1 Linear regression: one feature	50
	3.2 Linear regression: multiple features	51
	3.3 Categorical features	54
	3.4 Regularization	56
	3.5 Ridge regression	56
	3.6 Lasso regression	60
	3.7 Elastic Net regression	63
	3.8 Application to Predicting House Prices	64
	3.9 Logistic regression	69
	3.10 Decision criteria	71
	3.11 Application to credit decisions	72
	3.12 The k-nearest neighbor algorithm	79
	Summary	80
	Short concept questions	81
	Exercises	82
Chapter 4	Supervised Learning: Decision Trees	85
	4.1 Nature of decision trees	86
	4.2 Information gain measures	87
	4.3 Application to LendingClub data	89
	4.4 The naïve Bayes classifier	94
	4.5 Continuous target variables	98
	4.6 Ensemble learning	102
	Summary	103
	Short concept questions	104
	Exercises	105
Chapter 5	Supervised Learning: SVMs	107
	5.1 Linear SVM classification	107
	5.2 Modification for soft margin	114
	5.3 Non-linear separation	117
	5.4 Predicting a target’s value	119
	Summary	122
	Short concept questions	123
	Exercises	124
Chapter 6	Supervised Learning: Neural Networks	125
	6.1 ANNs with a single hidden layer	125
	6.2 Multi-layer ANNs	130
	6.3 Other activation functions	132
	6.4 Gradient descent algorithm	134
	6.5 Variations on the basic method	140
	6.6 The stopping rule	141
	6.7 The Black−Scholes−Merton formula	143
	6.8 Extensions	146
	6.9 Understanding volatility movements	147
	Summary	152
	Short concept questions	152
	Exercises	153
Chapter 7	Further Applications of Neural Networks	155
	7.1 Autoencoders and PCA	155
	7.2 General autoencoder design	158
	7.3 Variational autoencoders	160
	7.4 Generative adversarial networks	161
	7.5 Recurrent neural networks	163
	7.6 Convolutional neural networks	165
	Summary	169
	Short concept questions	169
	Exercises	170
Chapter 8	Reinforcement Learning	171
	8.1 The multi-armed bandit problem	172
	8.2 Changing environment	177
	8.3 The game of Nim	179
	8.4 Temporal difference learning	182
	8.5 When the opponent learns	185
	8.6 Deep Q-learning	186
	8.7 Playing chess	186
	8.8 Applications	187
	8.9 Optimal trade execution	189
	8.10 Data issues	192
	Summary	192
	Short concept questions	193
	Exercises	194
Chapter 9	Natural Language Processing	195
	9.1 Sources of data	198
	9.2 Pre-processing	199
	9.3 Bag-of-words model	200
	9.4 Application of naïve Bayes classifier	202
	9.5 Application of other algorithms	206
	9.6 Information retrieval	207
	9.7 Other NLP applications	208
	Summary	210
	Short concept questions	211
	Exercises	211
Chapter 10	Model Interpretability	213
	10.1 Linear regression	215
	10.2 Logistic regression	219
	10.3 Black-box models	222
	10.4 Shapley values	223
	10.5 LIME	226
	Summary	226
	Short concept questions	227
	Exercises	228
Chapter 11	Issues for Society	229
	11.1 Data privacy	230
	11.2 Biases	231
	11.3 Ethics	232
	11.4 Transparency	233
	11.5 Adversarial machine learning	233
	11.6 Legal issues	234
	11.7 Man vs. machine	235
Answers to End of Chapter Questions			237
Glossary of Terms			255
Index			265