# 【ML】机器学习论文与书籍推荐

## 基本模型

HMM (Hidden Markov Models，隐含马尔可夫模型)
[1] A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition

ME (Maximum Entropy，最大熵)
[2] A Maximum Entropy Approach to Natural Language Processing

MEMM (Maximum Entropy Markov Models，最大熵马尔可夫模型)
[3]  Maximum Entropy Markov Models for Information Extraction and Segmentation

CRF (Conditional Random Fields，条件随机场)
[4] An Introduction to Conditional Random Fields for Relational Learning
[5] Conditional Random Fields – Probabilistic Models for Segmenting and Labeling Sequence Data

SVM (Support Vector Machine，支持向量机)
[6] 统计学习理论 (张学工) # 缺

LSA/LSI (Latent Sematic Analysis / Indexing)
[7] An Introduction to Latent Semantic Analysis

pLSA/pLSI (Probablistic Latent Sematic Analysis / Indexing)
[8] Probabilistic Latent Semantic Analysis

LDA(Latent Dirichlet Allocation)
[9] Latent Dirichlet Allocation # 用变分理论和最大化期望算法求解模型
[10] Parameter estimation for text analysis # 用吉布斯采样求解模型

Neural Networks (神经网络，包括霍普菲尔模型、自组织映射、随机网络、玻尔兹曼机等)
[11] Neural Networks – A Systematic Introduction

Diffusion Networks (扩散网络)
[12] Diffusion Networks, Products of Experts, and Factor Analysis

Markov Random Fields (马尔可夫随机场)

Generalized Linear Model (广义线性模型，包括逻辑回归等)
[13] An introduction to Generalized Linear Models (2nd) # 有第三版，不过我没寻到

Chinese Restaurant Model (中餐馆模型？，狄利克雷过程)
[14] Dirichlet Processes, Chinese Restaurant Processes and all that
[15] Estimating a Dirichlet Distribution

## 一些重要算法

EM (Expectation Maximization，期望值最大化)
[16] Expectation Maximization and Posterior Constraints
[17] Maximum Likelihood from Incomplete Data via the EM Algorithm

MCMC & Gibbs Sampling (马尔可夫链蒙特卡罗算法与吉布斯采样)
[18] Markov Chain Monte Carlo and Gibbs Sampling
[19] Explaining the Gibbs Sampler
[20] An Introduction to MCMC for Machine Learning

PageRank

SVD、QR 分解、Shur 分解、LU 分解、谱分解

Spectral Clustering (谱聚类)
[22] A Tutorial on Spectral Clustering

Energy-Based Learning
[23] A Tutorial on Energy-Based Learning

Belief Propagation (置信传播)
[24] Understanding Belief Propagation and its Generalizations
[25] Construction free energy approximation and generalized belief propagation algorithms
[26] Loopy Belief Propagation for Approximate Inference An Empirical Study
[27] Loopy Belief Propagation

AP (Affinity Propagation，亲缘传播)
[28] Affinity Propagation

L-BFGS
[29] 最优化理论与算法(第二版) 第十章 # 缺
[30] On the Limited Memory BFGS Method for Large Scale Optimization

IIS (Improved Iterative Scaling，改进迭代算法)
[31] Improved Iterative Scaling Algorithm – Parameter Estimation of Feature-based Model

## 理论部分

Probabilistic Networks (概率网络)
[32] An Introduction to Variational Methods
[33] Factor Graphs and the Sum-Product Algorithm
[34] Constructing Free Energy Approximations and Generalized Belief Propagation Algorithms
[35] Graphical Models, Exponential Families, and Variational Inference

Variational Theory (变分理论) # 我们只用概率网络上的变分
[36] Tutorial on variational approximation methods
[37] A Variational Bayesian Framework for Graphical Models
[38] Tutorial on variational approximation methods (ppt) # 原文为 variational tutorial.pdf，未寻到

Information Theory (信息论)
[39] Elements of Information Theory (2nd)

[40] 测度论(Halmos) # 缺，据说写得好但是有点过时了
[41] 测度论讲义(严加安)

[42] 概率与测度论 # 缺

[43] 应用随机过程(林元烈) # 缺
[44] 随机数学引论(林元烈) # 缺

Matrix Theory (矩阵论)
[45] 矩阵分析与应用(张贤达) # 缺

[46] 模式识别 (2nd)(边肇祺) # 缺
[47] Pattern Recognition and Machine Learning

[48] Convex Optimization
[49] 最优化理论与算法(陈宝林)

[50] 泛函分析导论及应用

[51] 模式分析的核方法 # 缺

[52] 统计手册 # 缺

## 综合部分

Semi-Supervised Learning (半监督学习)
[53] Semi-Supervised Learning (MIT Press) # 缺
[54] Graph-Based Semi-Supervised Learning

Co-Training (协同训练)

Self-Training (自我训练)