机器学习算法(第2版影印版)(英文版)

机器学习因运用大数据实现强大且快速的预测而 大受欢迎。然而，其强大的输出背后，真正力量来自 复杂的算法，涉及大量的统计分析，以大数据作为驱 动而产生实质性的洞察力。本书第2版的机器学习算法 引导您取得与机器学习过程中的主要算法相关的显著 开发结果，并帮助您加强和掌握有监督，半监督和加 强学习等领域的统计解释。一旦全面吃透了算法的核 心概念，您将基于最广泛的库（如sclkit-learn、 NLTK、TensorFlow和Keras）来探索现实世界的示例。 您将发现新的主题，如主成分分析（PCA）、独立成分 分析（ICA）、贝叶斯回归、判别分析、高级聚类和高 斯混合等。

Preface Chapter 1: A Gentle Introduction to Machine Learning Introduction - classic and adaptive machines Descriptive analysis Predictive analysis Only learning matters Supervised learning Unsupervised learning Semi-supervised learning Reinforcement learning Computational neuroscience Beyond machine learning - deep learning and bio-inspired adaptive systems Machine learning and big data Summary Chapter 2: Important Elements in Machine Learning Data formats Multiclass strategies One-vs-all One-vs-one Learnability Underfitting and overfitting Error measures and cost functions PAC learning Introduction to statistical learning concepts MAP learning Maximum likelihood learning Class balancing Resampling with replacement SMOTE resampling Elements of information theory Entropy Cross-entropy and mutual information Divergence measures between two probability distributions Summary Chapter 3: Feature Selection and Feature Engineering scikit-learn toy datasets Creating training and test sets Managing categorical data Managing missing features Data scaling and normalization Whitening Feature selection and filtering Principal Component Analysis Non-Negative Matrix Factorization Sparse PCA Kernel PCA Independent Component Analysis Atom extraction and dictionary learning Visualizing high-dimensional datasets using t-SNE