Instructors: Seid Muhie Yimam
Event type:
Seminar
Displayed in timetable as:
Sem Data science
Hours per week:
2
Credits:
3,0
Language of instruction:
English
Min. | Max. participants:
- | 20
Registration group: Anmeldegruppe Seminare
Comments/contents:
The integration of machine learning models into different applications, including in business and finance, legal domain, healthcare, and humanities is increasing over time. The biases introduced in the data could lead to catastrophic interpretation mistakes. In this course, we will discuss how ML models can be interpreted and explained, hence studying the details of interpretable and explainable models. We will use real-world examples, apply basic data curation strategies, and amend models on the way.
Learning objectives:
The seminar provides theoretical and empirical insights into the relationship between digitalization, society, and politics. It will enable the participants to take part in profound discussions about the preconditions and implications of digital networks, business models as well as technical and legal regulations.
Didactic concept:
The seminar will take place in person.
The participants will work on the seminar topics through presentations by the lecturer, provided literature, and their own research, which they will present in class. Regular attendance of the lecture is expected.
Literature:
- Serg Masîs, 2021, Interpretable Machine Learning with Python: Learn to Build Interpretable High-Performance Models with Hands-on Real-world Examples
Additional examination information:
Weekly/bi-weekly presentation of seminars on selected topics, Group work on a selected project on the interpretable and explainable model, and data curation.
|