Supervised Learning

  1. Support Vector Machines
    1. Linear SVM
      1. The Concept of a Hyperplane
        1. Separating Hyperplane
          1. Mathematical Representation
          2. Maximizing the Margin
            1. Margin Definition
              1. Support Vectors
                1. Optimal Hyperplane
                2. Hard Margin SVM
                  1. Linearly Separable Data
                    1. Optimization Problem Formulation
                      1. Lagrangian Formulation
                      2. Soft Margin SVM
                        1. Handling Non-separable Data
                          1. Slack Variables
                            1. Regularization Parameter C
                            2. Dual Formulation
                              1. Quadratic Programming Problem
                                1. Support Vector Identification
                              2. Non-linear SVM
                                1. The Kernel Trick
                                  1. Feature Space Transformation
                                    1. Kernel Function Properties
                                    2. Common Kernel Functions
                                      1. Linear Kernel
                                        1. Polynomial Kernel
                                          1. Degree Parameter
                                            1. Coefficient Parameters
                                            2. Radial Basis Function Kernel
                                              1. Gaussian RBF
                                                1. Gamma Parameter
                                                2. Sigmoid Kernel
                                                  1. Custom Kernels
                                                  2. Kernel Selection
                                                    1. Problem-specific Considerations
                                                      1. Cross-validation for Kernel Choice
                                                    2. SVM for Classification
                                                      1. Binary Classification
                                                        1. Multiclass Classification
                                                          1. One-vs-One Strategy
                                                            1. One-vs-Rest Strategy
                                                              1. Direct Multiclass Methods
                                                              2. Decision Function
                                                                1. Probability Estimates
                                                                2. Support Vector Regression
                                                                  1. Epsilon-Insensitive Loss
                                                                    1. SVR Formulation
                                                                      1. Kernel Methods in Regression
                                                                        1. Hyperparameter Selection
                                                                        2. Practical Considerations
                                                                          1. Feature Scaling Requirements
                                                                            1. Computational Complexity
                                                                              1. Memory Requirements
                                                                                1. Parameter Tuning Strategies

                                                                              Previous

                                                                              5. Instance-Based Learning

                                                                              Go to top

                                                                              Next

                                                                              7. Probabilistic Models