比较系统的在线学习网站

转自机器学习研究会

这个网站整理了计算机相关的知识，包括数据结构与算法、统计、机器学习、最优化等等。点击by topic可以看到目录（下图是机器学习的目录），每个concept也列出了先修知识。推荐一下~

http://metacademy.org/

Machine Learning

Roadmaps

• Bayesian machine learning

• Deep learning from the bottom up

• Differential geometry for machine learning

• Level-Up Your Machine Learning

Course Guides

• Berkeley CS281a: Statistical Learning Theory

• Coursera: Machine Learning

• Stanford CS229: Machine Learning

Concepts

• AdaBoost

• Akaike information criterion

• annealed importance sampling

• backpropagation

• backpropagation for second-order methods

• bagging

• basis function expansions

• Baum-Welch algorithm

• Bayes net parameter learning

• Bayes net structure learning

• Bayes' rule

• Bayesian decision theory

• Bayesian linear regression

• Bayesian logistic regression

• Bayesian model averaging

• Bayesian model comparison

• Bayesian naive Bayes

• Bayesian networks

• Bayesian parameter estimation

• Bayesian parameter estimation in exponential families

• Bayesian parameter estimation: Gaussian distribution

• Bayesian parameter estimation: multinomial distribution

• Bayesian parameter estimation: multivariate Gaussians

• Bayesian PCA

• beta process

• bias-variance decomposition

• binary linear classifiers

• Boltzmann machines

• boosting as optimization

• Chinese restaurant franchise

• Chinese restaurant process

• Chow-Liu trees

• collapsed Gibbs sampling

• comparing Gaussian mixtures and k-means

• computations on multivariate Gaussians

• conditional random fields

• constructing kernels

• convolutional neural nets

• cross validation

• CRP clustering

• curse of dimensionality

• decision trees

• deep belief networks

• early stopping

• EM algorithm for PCA

• Expectation-Maximization algorithm

• F measure

• factor analysis

• feed-forward neural nets

• Fisher's linear discriminant

• fitting logistic regression with iterative reweighted least squares

• forward-backward algorithm

• gamma distribution

• Gaussian discriminant analysis

• Gaussian process classification

• Gaussian process regression

• Gaussian processes

• generalization

• generalized linear models

• Gibbs sampling

• Gibbs sampling as a special case of Metropolis-Hastings

• GP classification with the Laplace approximation

• Hamiltonian Monte Carlo

• hidden Markov models

• hierarchical Dirichlet process

• Hopfield networks

• IBP linear-Gaussian model

• independent component analysis

• Indian buffet process

• information form for multivariate Gaussians

• Jensen's inequality

• K nearest neighbors

• k-means

• k-means++

• kernel SVM

• kernel trick

• Laplace approximation

• LASSO

• latent Dirichlet allocation

• latent semantic analysis

• learning Bayes net parameters with missing data

• learning GP hyperparameters

• learning invariances in neural nets

• learning linear dynamical systems

• linear regression

• linear regression as maximum likelihood

• linear regression: closed-form solution

• linear-Gaussian models

• logistic regression

• MAP parameter estimation

• Markov chain Monte Carlo

• Markov chains

• Markov models

• Markov random fields

• maximum likelihood

• maximum likelihood in exponential families

• MCMC convergence

• mean field approximation

• Metropolis-Hastings algorithm

• mixture of Gaussians models

• MRF parameter learning

• multidimensional scaling

• multinomial logistic regression

• naive Bayes

• perceptron algorithm

• precision and recall

• principal component analysis

• principal component analysis (proof)

• probabilistic Latent Semantic Analysis

• probabilistic PCA

• probit regression

• random forests

• recurrent neural networks

• restricted Boltzmann machines

• reversible jump MCMC

• ridge regression

• ridge regression as SVD

• sequential Monte Carlo

• slice sampling

• soft margin SVM

• soft weight sharing in neural nets

• sparse coding

• structured mean field

• support vector machine

• support vector regression

• SVM optimality conditions

• SVM vs. logistic regression

• tangent propagation

• unsupervised pre-training

• variational Bayes

• variational Bayes EM

• variational inference

• variational inference and exponential families

• variational linear regression

• variational logistic regression

• variational mixture of Gaussians

• VC dimension

• Viterbi algorithm

• weight decay in neural networks