机器学习技法

线性支持向量机、对偶支持向量机、核型支持向量机、软式支持向量机、核逻辑回归、支持向量回归

• 课时1：Course Introduction (4分7秒)

• 课时2：Large-Margin Separating Hyperplane (14分17秒)

• 课时3：Standard Large-Margin Problem (19分16秒)

• 课时4：Support Vector Machine (15分33秒)

• 课时5：Reasons behind Large-Margin Hyperplane (13分31秒)

• 课时6：Motivation of Dual SVM (15分54秒)

• 课时7：Lagrange Dual SVM (18分50秒)

• 课时8：Solving Dual SVM (14分19秒)

• 课时9：Messages behind Dual SVM (11分18秒)

• 课时10：Kernel Trick (20分23秒)

• 课时11：Polynomial Kernel (12分16秒)

• 课时12：Gaussian Kernel (14分43秒)

• 课时13：Comparison of Kernels (13分35秒)

• 课时14：Motivation and Primal Problem (14分28秒)

• 课时15：Dual Problem (7分38秒)

• 课时16：Messages behind Soft-Margin SVM (13分44秒)

• 课时17：Model Selection (9分57秒)

• 课时18：Soft-Margin SVM as Regularized Model (13分40秒)

• 课时19：SVM versus Logistic Regression (10分18秒)

• 课时20：SVM for Soft Binary Classification (9分37秒)

• 课时21：Kernel Logistic Regression (16分22秒)

• 课时22：Kernel Ridge Regression (17分17秒)

• 课时23：Support Vector Regression Primal (18分44秒)

• 课时24：Support Vector Regression Dual (13分5秒)

• 课时25：Summary of Kernel Models (9分6秒)

• 课时26：Motivation of Aggregation (18分54秒)

• 课时27：Uniform Blending (20分31秒)

• 课时28：Linear and Any Blending (16分48秒)

• 课时29：Bagging (Bootstrap Aggregation) (11分48秒)

• 课时30：Motivation of Boosting (12分47秒)

• 课时31：Diversity by Re-weighting (14分28秒)

• 课时32：Adaptive Boosting Algorithm (13分34秒)

• 课时33：Adaptive Boosting in Action (11分5秒)

• 课时34：Decision Tree Hypothesis (17分28秒)

• 课时35：Decision Tree Algorithm (15分20秒)

• 课时36：Decision Tree Heuristics in CRT (13分21秒)

• 课时37：Decision Tree in Action (8分44秒)

• 课时38：Random Forest Algorithm (13分6秒)

• 课时39：Out-Of-Bag Estimate (12分32秒)

• 课时40：Feature Selection (19分27秒)

• 课时41：Random Forest in Action (13分28秒)

• 课时42：Adaptive Boosted Decision Tree (15分6秒)

• 课时43：Optimization View of AdaBoost (27分25秒)

• 课时44：Gradient Boosting (18分20秒)

• 课时45：Summary of Aggregation Models (11分19秒)

• 课时46：Motivation (20分36秒)

• 课时47：Neural Network Hypothesis (18分1秒)

• 课时48：Neural Network Learning (20分15秒)

• 课时49：Optimization and Regularization (19分29秒)

• 课时50：Deep Neural Network (21分30秒)

• 课时51：Autoencoder (15分17秒)

• 课时52：Denoising Autoencoder (8分31秒)

• 课时53：Principal Component Analysis (31分20秒)

• 课时54：RBF Network Hypothesis (12分55秒)

• 课时55：RBF Network Learning (20分8秒)

• 课时56：k-Means Algorithm (16分19秒)

• 课时57：k-Means and RBF Network in Action (9分46秒)

• 课时58：Linear Network Hypothesis (20分16秒)

• 课时59：Basic Matrix Factorization (16分32秒)

• 课时60：Stochastic Gradient Descent (12分22秒)

• 课时61：Summary of Extraction Models (9分12秒)

• 课时62：Feature Exploitation Techniques (16分11秒)

• 课时63：Error Optimization Techniques (8分40秒)

• 课时64：Overfitting Elimination Techniques (6分44秒)

• 课时65：Machine Learning in Action (12分59秒)