Content of the lecture

Remarkable progress has been made in the field of artificial intelligence (AI) in recent years. Machine learning (ML) is a sub-discipline of AI that attempts to develop techniques that enable computers to learn from data. The goal of ML methods is to reliably abstract the underlying model for specific tasks.

In the lecture Practical Machine Learning (PML), the theoretical foundations as well as the basic concepts and techniques of machine learning are covered, but the focus is on problem solving and practical application. The course offers the opportunity to explore different ML algorithms and their applications in various fields, including computer vision, natural language processing and data mining. During the course, you will have the opportunity to work on various application tasks and a group project in which you will apply the concepts you have learned to real-world datasets. You will learn how to use common libraries and tools for ML such as Scikit-Learn, TensorFlow and Keras and apply them to real-world datasets. You will also learn how to evaluate the performance of your models and interpret their results.

The lecture style will be a mix of theory and practical applications, with an emphasis on problem solving and hands-on experimentation. The theoretical part of the lecture will be offered as a block course at the beginning of the semester (early/mid April). Students then have the opportunity to work on a problem from the field of ML alone or in small groups during the semester and present their results in the form of a scientific essay. The quality assurance of the essay is carried out through a mutual peer review process, in which the students benefit from mutual feedback both from a technical point of view and with regard to the presentation of content.

The module covers the fundamentals and concepts of machine learning. Topics covered include the following:

• Introduction to machine learning and its applications.
• Data pre-processing and feature engineering techniques.
• Supervised and unsupervised learning algorithms.
• Deep learning techniques such as Convolutional Neural Networks and Recurrent Neural Networks.
• Transfer learning and Tiny ML.
• Evaluation metrics for ML models.
• Hyperparameter tuning and model selection techniques.
• Interpreting the results of ML models.

... other interesting topics.

Information on the examination

Success is assessed by the submission of the project work and a 30-minute presentation of the project work.