Methods of AI – Machine Learning
This course explains machine learning using decision trees and neural networks – two fundamental learning algorithms used for supervised machine learning.
Overview
This course introduces two very relevant learning algorithms used for supervised learning: decision trees, which lie beneath modern boosting methods, and neural networks, which are at the heart of deep learning.
The course explains how to apply these algorithms to practical problems. Best practices for training and evaluation are discussed. Quality metrics such as accuracy, precision, and recall are explained. The course concludes with a discussion of the risks and challenges of neural networks.
This course is based on chapters 19 and 21 of the textbook by Stuart Russell and Peter Norvig: Introduction to Artificial Intelligence - A Modern Approach.
The course is held in English language with German subtitles.
Which topics are covered?
- Types of learning (unsupervised, supervised, reinforcement)
- Decision trees
- Information gain and entropy
- Neural Networks
- "Unit" model of a neuron McCulloch and Pitts
- Activation functions
- Backpropagation algorithm
- Machine learning workflow
- Training and test data
- K-fold cross validation
- Accuracy, sensitivity, specificity, precision, F1 score
What will I achieve?
At the end of the course, you will be able to
- understand and apply machine learning using decision trees and neural networks,
- evaluate a trained classifier,
- understand the challenges and risks of neural networks.
What requirements do I have to meet?
None.
