2D1431 Machine Learning Lab 2: Bayes Classifier & Boosting

In this lab you will implement a Bayes Classifier and the Adaboost algorithm that improves the performance of a weak classifier by aggregating multiple hypotheses generated across different distributions of the training data. Some predefined functions for visualization and basic operations are provided, but you will have to program the key algorithms yourself. During the examination with the lab assistant, you will present the results, answers to the questions and the code that you wrote. It is assumed that you are familiar with the basic concepts of Bayesian learning and that you have read chapter 6 in the course book Machine Learning (Mitchell, 1997) and the papers by Schapire (1999) and Quinlan (1996) on boosting. If you look for additional literature, chapters 2, 3 and 9 in the book by Duda et al. (2001) are worthwhile reading. In this exercise we will work with two images. Start by looking at the images.