Lectures, workgroups and supervised computer practicals.
Students will need to have basic knowledge of calculus, probability theory, lineair algebra and possess basic programming skills.
Upon completion of the course, students will:
- Have gained an overview of artificial neural networks.
- Be able to place the domain of neural networks in a historical context and conclude different principles underlying classic and modern artificial neural networks.
- Be able to describe behaviour of different models in terms of formal properties.
- Be able to implement various neural network models in Python.
During the lectures, the formal concepts underlying modern neural networks will be developed, including deep neural networks and recurrent neural networks. Also, various classic neural network models will be discussed like the (Multi-layer) Perceptron, Hopfield networks and Boltzmann machines. During the wrokgroups, students will get acquainted with the theoretical background of neural networks. During the practicals, students will get to immers themselves in the theoretical and practical aspects of neural networks. Each student will get to implement various models using Python, a programming language which they will learn to use during the course.
Workgroup and practical assignments.
The course will continue into the fourth period for a number of weeks.