数学似宇宙，韭菜只关心其中实用的部分。
scikit-learn (sklearn) 官方文档中文版
scikit-learn Machine Learning in Python
一个新颖的online图书资源集，非常棒。
Bayesian Machine Learning
9. 【ignore】
随机过程
[Scikit-learn] 1.1 Generalized Linear Models - Bayesian Ridge Regression【等价效果】
8. [Bayesian] “我是bayesian我怕谁”系列 - Variational Autoencoders
稀疏表达
7. 【ignore】
贝叶斯网络
[Scikit-learn] Dynamic Bayesian Network - Conditional Random Field【去噪、词性标注】
6. 【隐马及其扩展】
时序模型
概率降维
[Scikit-learn] 4.4 Dimensionality reduction - PCA
[Scikit-learn] 2.5 Dimensionality reduction - Probabilistic PCA & Factor Analysis
[Scikit-learn] 2.5 Dimensionality reduction - ICA
[Scikit-learn] 1.2 Dimensionality reduction - Linear and Quadratic Discriminant Analysis
4. 【公式推导解读】
概率聚类
[Scikit-learn] 2.1 Clustering - Gaussian mixture models & EM
[Scikit-learn] 2.1 Clustering - Variational Bayesian Gaussian Mixture
3. 【概念解读】
隐变量模型
[Bayes] Concept Search and LSI
[Bayes] Concept Search and PLSA
[Bayes] Concept Search and LDA
2. 【ignore】
朴素贝叶斯
常见分布关系
<Statistical Inference> goto: 647/686
先验分布与后验分布
其中两个概念比较重要：
-
-
- 无信息先验分布 (Non-informative prior)
- Jeffreys先验分布 (Jeffreys prior)
-
后验即是：贝叶斯统计推断
-
-
- 后验分布与充分性 (Posterior distribution and sufficiency)
- 无信息先验下的后验分布 (Posterior distribution with noninformative prior)
- 共轭先验下的后验分布 (Posterior distribution with conjugate prior)
-
结合损失函数：贝叶斯统计决策
-
-
- 平方损失 (square loss)
- 加权平方损失 (weighted squared loss)
- 绝对值损失 (absolute loss)
- 线性损失函数 (linear loss function)
-
抽样方法
一种逼近求值策略：贝叶斯计算方法
-
- 【采样法大纲】
- 直接抽样法 & 可视化方法
-
- 接受-拒绝抽样（Acceptance-Rejection sampling)
- 重要性抽样（Importance sampling）
-
- MCMC抽样方法
(a). Metropolis-Hasting算法
(b). Gibbs采样算法
-
- 采样估参
其他未整理
non-Bayesian Machine Learning
Algorithm Outline
基本概念
基本算法
[Scikit-learn] 1.5 Generalized Linear Models - SGD for Regression
[Scikit-learn] 1.5 Generalized Linear Models - SGD for Classification
Online Learning
[Scikit-learn] 1.1 Generalized Linear Models - Comparing various online solvers
[Scikit-learn] Yield miniBatch for online learning.
线性问题
[UFLDL] Linear Regression & Classification
线性拟合
[Scikit-learn] Theil-Sen Regression【抗噪能力较好】
线性分类
# Discriminative Models
# Generative Models
Naive Bayes【参见 "贝叶斯机器学习"】
决策树
降维
聚类
[Scikit-learn] 2.3 Clustering - kmeans
[Scikit-learn] 2.3 Clustering - Spectral clustering
[Scikit-learn] *2.3 Clustering - DBSCAN: Density-Based Spatial Clustering of Applications with Noise
[Scikit-learn] *2.3 Clustering - MeanShift
End.