Introduction to Machine Learning Part 3: Evaluation and Tuning

Machine Learning (ML) is at the core of many applications of artificial intelligence. A key goal of this course series is to teach the fundamental building blocks behind supervised ML. In this third (and last) part we will cover the evaluation and tuning of ML models in depth. In addition to understanding the theoretical basics, participants will learn how to apply and evaluate different ML algorithms in R and Python with a focus on tabular data.

• Evaluation of classification models via, e.g., ROC metrics and AUC
• Evaluation of regression models
• Hyperparameter tuning
• Nested resampling
   

• Explain how a classification model can be evaluated using ROC metrics and the AUC
• Explain the differences between resampling strategies such as cross-validation and bootstrap
• Explain the importance of hyperparameter tuning and know the basic algorithms for doing that
• Explain the concepts of overfitting and overtuning

This course is open to all who are interested. However, we recommend learners to have:

• A strong foundation in mathematics, such as 8 years of math education in secondary schools
• Pre-knowledge in linear algebra and analysis required (at least high school level)
• Pre-knowledge in statistics and probability is recommended (at least high school level)
• Basic programming skills in R or Python (e.g., through a small self-study course)
• You have concluded the following courses: Introduction to Machine Learning Part 1 & Introduction to Machine Learning Part 2