A practical guide to dimensionality reduction

This 4-hour course provides an introduction to the topic of dimensionality reduction and serves as a starting point for self-guided learning during and beyond the course time.

The course covers alternating sequences of theoretical input and hands-on exercises, which are discussed with the instructors during the course.

Dimensionality reduction is a common data preprocessing step preceding the application of supervised and unsupervised learning methods in AI modeling. After motivating the use of dimensionality reduction and highlighting its role in data exploration, this course gives an introduction to three types of dimensionality reduction approaches: feature transformation, feature aggregation, and feature selection. Course participants will have the opportunity to discover and compare the main methods for each approach in a hands-on experience, using jupyter notebooks on a real-world high-dimensional dataset.

Learning goals

Day 1: General introduction and feature transformation methods

• General introduction to dimensionality reduction
• Theory and practical application of classical feature transformation methods
• Theory and practical application of autoencoders for feature transformation

Day 2: Further unsupervised and supervised methods

• Theory and practical application of feature aggregation approaches
• Theory and practical application of feature selection methods
• Stability optimization in feature selection

Prerequisites

Basic knowledge of 

• Python, see the courses
  • “First steps in Python” 
  • “Data processing with Pandas & Data visualization with Matplotlib”
• ML models, see the courses
  • "Introduction to Machine Learning"
  • "Basic Methods of Machine Learning"

Google account is recommended.

Target group

This course is open to researchers of all career stages, or anyone interested in learning about the subject.

This course is free of charge.