Machine Learning Overview

All machine learning models, wether they are probabilistic or non-probabilistic, parametric or non-parametric, generative or disciminative have in common that they are trained on some sort of trainig data. This seperates them from rule-based systems where the developer explicitly models his knowledge leading to specific models. This qualification line introduces the basic mathematical concepts and gives a broad overview of the most common machine learning models.


Bayesian Classification

[slides] Histograms for bayesian classification

[notebook] KDE and KNN with Python

[slides] Modelling of Priors

Decision Trees and Random Forests

[slides] Decision Tree

[slides] Random Forest

[notebook] Decision Tree and Random Forest with Python

Logistic Regression

[slides] Logistic Regression

[notebook] Logistic Regression with Python using Scikit-Learn

Support Vector Machine

[slides] Support Vector Machines

[notebook] Support Vector Machine with Python using Scikit-Learn

Neural Networks

[slides] Neural Networks - Basics

Additional Topics

[slides] Ensambles and Boosting