Machine Learning (ML) is an umbrella concept to computational algorithms, aiming at solving (large scale) problems without precise instructions on how to find solutions. The different algorithms (e.g., Perceptron, Self-Organizing Maps, Support Vector Machines, Random Forests, Autoencoders, Transformers, etc.) have in common that they can radically upscale otherwise relatively simple statistical or mathematical concepts. It is, however, important to understand the inner workings of ML and to interpret the results according to exactly how the problem is formulated.  A common way to set up ML experiments is to use existing libraries, such as TensorFlow, and configure the algorithms using the Python programming language. Python is thus a necessary tool to perform experiments.     

The following topics are covered: 

• Basic statistical concepts and terminology such as sampling, variation, probability, modeling, inference, etc 
• Read in data in various file formats such as Excel, comma-separated text, 
• Data manipulation, graphics, use of the R environment for descriptive and exploratory data analysis. 
• Null Hypothesis, scientific testing, z-values of the normal distribution, t-distribution and t-tests, multiple testing 
• Linear models, OLS, logistic regression, residuals, model validation and model selection (multiple regression) 

ML topics covered include: 

• The Linux/Unix command line shell (refresher) 
• Basic programmeming concepts  
• Programmeming in Python 
• Algorithms in ML and their applications 
• Using ML libraries in Python 
• Parsing and loading data 
• Processing data using ML algorithms