机器学习及其应用(英文版)

目录1. Introduction 引言… ………………………………………………………………………………11.1 Towards Intelligent Machines 走向智能机器 ………………………………………………11.2 Well-Posed Machine Learning Problems 适定的机器学习问题 ……………………………51.3 Examples of Applications in Diverse Fields 不同领域的应用实例 …………………………71.4 Data Representation 数据表示 ………………………………………………………………121.4.1 Time Series Forecasting 时间序列预测 ………………………………………………151.4.2 Datasets for Toy (Unreastically Simple) and Realistic Problems初级问题和现实问题数据集 …………………………………………………………171.5 Domain Knowledge for Productive use of Machine Learning使机器学习有效应用的领域知识 ……………………………………………………………181.6 Diversity of Data: Structured/Unstructured 数据多样性：结构化/非结构化 ……………201.7 Forms of Learning 学习形式 …………………………………………………………………211.7.1 Supervised/Directed Learning 监督/指导学习 ……………………………………211.7.2 Unsupervised/Undirected Learning 非监督/无指导学习 …………………………221.7.3 Reinforcement Learning 强化学习 …………………………………………………221.7.4 Learning Based on Natural Processes: Evolution, Swarming, and Immune Systems基于自然处理的学习：进化、集群和免疫系统………………………………………231.8 Machine Learning and Data Mining 机器学习和数据挖掘 …………………………………251.9 Basic Linear Algebra in Machine Learning Techniques机器学习技术中的基础线性代数 ……………………………………………………………261.10 Relevant Resources for Machine Learning 机器学习的相关资源 …………………………342. Supervised Learning: Rationale and Basics 监督学习：基本原理和基础… ………………362.1 Learning from Observations 从观测中学习 …………………………………………………362.2 Bias and Variance 偏差和方差 ………………………………………………………………422.3 Why Learning Works: Computational Learning Theory学习为什么有效：计算学习理论 ……………………………………………………………462.4 Occam's Razor Principle and Overfitting Avoidance奥卡姆剃刀原理和防止过拟合 ………………………………………………………………492.5 Heuristic Search in Inductive Learning 归纳学习中的启发式搜索 …………………………512.5.1 Search through Hypothesis Space 假设空间搜索 ……………………………………522.5.2 Ensemble Learning 集成学习 ………………………………………………………532.5.3 Evaluation of a Learning System 学习系统的评价 …………………………………552.6 Estimating Generalization Errors 估计泛化误差 ……………………………………………562.6.1 Holdout Method and Random Subsampling 留出法和随机下采样 …………………562.6.2 Cross-validation 交叉验证 ……………………………………………………………572.6.3 Bootstrapping 自助法 …………………………………………………………………582.7 Metrics for Assessing Regression (Numeric Prediction) Accuracy评价回归（数值预测）精度的指标 …………………………………………………………592.7.1 Mean Square Error 均方误差 …………………………………………………………602.7.2 Mean Absolute Error 平均绝对误差 …………………………………………………602.8 Metrics for Assessing Classification (Pattern Recognition) Accuracy评价分类（模式识别）精度的指标 …………………………………………………………612.8.1 Misclassification Error 误分类误差 …………………………………………………612.8.2 Confusion Matrix 混淆矩阵 …………………………………………………………622.8.3 Comparing Classifiers Based on ROC Curves 基于ROC曲线的分类器比较 ……662.9 An Overview of the Design Cycle and Issues in Machine Learning机器学习中的设计周期和问题概述 …………………………………………………………683. Statistical Learning 统计学习… …………………………………………………………………733.1 Machine Learning and Inferential Statistical Analysis 机器学习与推断统计分析 …………733.2 Descriptive Statistics in Learning Techniques 学习技术中的描述统计学 …………………743.2.1 Representing Uncertainties in Data: Probability Distributions表示数据中的不确定性：概率分布 …………………………………………………753.2.2 Descriptive Measures of Probability Distributions 概率分布的描述方法 …………803.2.3 Descriptive Measures from Data Sample 数据样本的描述方法 ……………………833.2.4 Normal Distributions 正态分布 ………………………………………………………843.2.5 Data Similarity 数据相似性 …………………………………………………………853.3 Bayesian Reasoning: A Probabilistic Approach to Inference贝叶斯推理：推断的概率方法 ………………………………………………………………873.3.1 Bayes Theorem 贝叶斯定理 …………………………………………………………883.3.2 Naive Bayes Classifier 朴素贝叶斯分类器 …………………………………………933.3.3 Bayesian Belief Networks 贝叶斯信念网络 …………………………………………983.4 k-Nearest Neighbor (k-NN) Classifier k近邻（k-NN）分类器 …………………………1023.5 Discriminant Functions and Regression Functions 判别函数和回归函数 ………………1063.5.1 Classification and Discriminant Functions 分类和判别函数 ……………………1073.5.2 Numeric Prediction and Regression Functions 数值预测和回归函数 ……………1083.5.3 Practical Hypothesis Functions 实践应用中的假设函数 …………………………1093.6 Linear Regression with Least Square Error Criterion基于*小二乘误差准则的线性回归法 ……………………………………………………1123.6.1 Minimal Sum-of-Error-Squares and the Pseudoinverse *小误差平方和与伪逆 1133.6.2 Gradient Descent Optimization Schemes 梯度下降法优化方案 …………………1153.6.