The Bayes classifier is a Data Scientist’s unbeatable arch enemy. When faced with the challenge of finding an optimal solution to a classification problem, it is the ultimate (theoretical!) benchmark when evaluating the performance of a classifier. The problem with the Bayes classifier is that it uses information that generally is not available to us …

# Category Archives: probability theory

## Means and medians

Understanding random experiments through deterministic invariants is a central goal of the mathematical branch known as probability theory or stochastics. While the expected value (or the mean) of a random experiment has a very natural interpretation (as its name suggests, it is the value we expect to occur on average when repeating the experiment) and …