Courses in this compulsory block cover essential topics of machine learning and also allow you to fully delve into its subclasses, deep learning and deep reinforcement learning.
In Introduction to Machine Learning with Python, Data Mining and Pattern Recognition and Machine Learning you will encounter, for example, simple (but still trendy) decision trees, clustering, well-known linear and logistic regression, and the basics of artificial neural networks. In Deep Learning, the focus is on deep neural networks, explaining their basics but also revealing the current critical tricks of the trade, the modern alchemy. Machine learning, deep or not, makes use of real-world data to build (or train) an algorithm that helps us to understand or solve a given problem. Deep Reinforcement Learning is targeted to situations where the reward comes much later than with each training example, thereby opening up many new applications in domains such as healthcare, robotics, smart grids, finance, and more. Besides all this, you will gain hands-on experience dealing with real-world data and using current technologies (Python, R, MATLAB– the language depends on the course chosen, and TensorFlow).
(not formally necessary but strongly recommended)
Introductory courses in:
- calculus (as learning is usually an optimization problem),
- linear algebra (as the data are often tables and hence matrices),
- statistics and probability theory (as many say, machine learning is just applied statistics).