BADS WS19/20 No.7 Statistical Learning

In this lecture, we explore the foundations of statistical learning. After a brief recap of Bayes Theorem, we review logistic regression, which is the maybe most popular "machine learning" method in the industry. The core part of the class focuses on challenges in predictive modeling including the curse of dimensionality, overfitting and the bias-variance trade-off. We also take a brief look at regularization to protect against overfitting and how it is implemented in the scope of (logistic) regression.