This a two day course on the computational, statistical and probabilistic nature of machine learning. Core aspects will be covered, such that, the participants will get a vocabulary for working with problems where machine learning might be appropriate. A primer of statistics and probability will be covered in the beginning of the course so that at the end of the course, the participants should have an understanding of the core concepts of machine learning, which will help the participants in interpreting what the model is doing.

The first day of the course will focus on the Supervised learning in machine learning. Main emphasis will be in why some statistical problems might be more appropriate to approach in a supervised fashion, where the problem has labels associated with data. The examples will be illustrated via Neural Networks, Support Vector Machines and Ensemble methods. Model selection will also be covered.

The second day of the course will focus on unsupervised learning approaches. In applications where data might lack certain labels or have no labels at all, or the model is required to be adaptive, an unsupervised approach may be more beneficial. Parametric and nonparametric approaches to unsupervised learning will be considered.

The course will have programming work in Python. Basics of statistical and probability theory will be helpful.

Schedule:

Day 1
Course introduction
Supervised Learning
Neural Networks
Support Vector Machines
Ensemble methods
Model Selection

Day2
Unsupervised Learning
Parametric Unsupervised Learning
Nonparametric Unsupervised Learning
Combining Supervised Learning and Unsupervised Learning

For university students, you will receive a recommendation to receive 0.5ETCS credits from participating in the course.

For on-site participants, lunch and snacks will be included in the course fee.