基本模型:
1、HMM(Hidden Markov Models):
A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition.pdf
2、ME(Maximum Entropy):
ME_to_NLP.pdf
3、MEMM(Maximum Entropy Markov Models):
memm.pdf
4、CRF(Conditional Random Fields):
An Introduction to Conditional Random Fields for Relational Learning.pdf
Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data.pdf
5、SVM(support vector machine):
张学工<<统计学习理论>>
6、LSA(or LSI)(Latent Semantic Analysis):
Latent semantic analysis.pdf
7、pLSA(or pLSI)(Probablistic Latent Semantic Analysis):
Probabilistic Latent Semantic Analysis.pdf
8、LDA(Latent Dirichlet Allocation):
Latent Dirichlet Allocaton.pdf(用variational theory + EM算法解模型)
Parameter estimation for text analysis.pdf(using Gibbs Sampling 解模)
9、Neural Networksi(including Hopfield Model& self-organizing maps & Stochastic networks & Boltzmann Machine etc.):
Neural Networks – A Systematic Introduction
10、Diffusion Networks:
Diffusion Networks, Products of Experts, and Factor Analysis.pdf
11、Markov random fields:
12、Generalized Linear Model(including logistic regression etc.):
An introduction to Generalized Linear Models 2nd
13、Chinese Restraunt Model (Dirichlet Processes):
Dirichlet Processes, Chinese Restaurant Processes and all that.pdf
Estimating a Dirichlet Distribution.pdf
=================================================================
Some important algorithms:
1、EM(Expectation Maximization):
Expectation Maximization and Posterior Constraints.pdf
Maximum Likelihood from Incomplete Data via the EM Algorithm.pdf
2、MCMC(Markov Chain Monte Carlo) & Gibbs Sampling:
Markov Chain Monte Carlo and Gibbs Sampling.pdf
Explaining the Gibbs Sampler.pdf
An introduction to MCMC for Machine Learning.pdf
3、PageRank:
4、矩阵分解算法:
SVD, QR分解, Shur分解, LU分解, 谱分解
5、Boosting( including Adaboost):
adaboost_talk.pdf
6、Spectral Clustering:
Tutorial on spectral clustering.pdf
7、Energy-Based Learning:
A tutorial on Energy-based learning.pdf
8、Belief Propagation:
Understanding Belief Propagation and its Generalizations.pdf
bp.pdf
Construction free energy approximation and generalized belief
9、propagation algorithms.pdf
Loopy Belief Propagation for Approximate Inference An Empirical Study.pdf
Loopy Belief Propagation.pdf
10、AP (affinity Propagation):
11、L-BFGS:
<<最优化理论与算法 2nd>> chapter 10
On the limited memory BFGS method for large scale optimization.pdf
12、IIS:
IIS.pdf
=================================================================
理论部分:
1、概率图(probabilistic networks):
An introduction to Variational Methods for Graphical Models.pdf
Probabilistic Networks
Factor Graphs and the Sum-Product Algorithm.pdf
Constructing Free Energy Approximations and Generalized Belief Propagation Algorithms.pdf
*Graphical Models, exponential families, and variational inference.pdf
2、Variational Theory(变分理论,我们只用概率图上的变分):
Tutorial on varational approximation methods.pdf
A variational Bayesian framework for graphical models.pdf
variational tutorial.pdf
3、Information Theory:
Elements of Information Theory 2nd.pdf
4、测度论:
测度论(Halmos).pdf
测度论讲义(严加安).pdf
5、概率论:
……
<<概率与测度论>>
6、随机过程:
应用随机过程 林元烈 2002.pdf
<<随机数学引论>>
7、Matrix Theory:
矩阵分析与应用.pdf
8、模式识别:
<<模式识别 2nd>> 边肇祺
Pattern Recognition and Machine Learning.pdf
9、最优化理论:
<<Convex Optimization>>
<<最优化理论与算法>>
10、泛函分析:
<<泛函分析导论及应用>>
11、Kernel理论:
<<模式分析的核方法>>
12、统计学:
……
<<统计手册>>
==========================================================
综合:
1、semi-supervised learning:
<<Semi-supervised Learning>> MIT Press
semi-supervised learning based on Graph.pdf
2、Co-training:
3、Self-training: