|
Module Number:
| 14021
|
| Module Title: | Explainable Machine Learning |
| |
Erklärbares Maschinelles Lernen
|
| Department: |
Faculty 1 - Mathematics, Computer Science, Physics, Electrical Engineering and Information Technology
|
| Responsible Staff Member: | -
Prof. Dr. rer. biol. hum. Schneider, Erich
|
| Language of Teaching / Examination: | English |
| Duration: | 1 semester |
| Frequency of Offer: |
Every winter semester
|
| Credits: |
6
|
| Learning Outcome: | Students understand the interpretability and explainability of machine learning systems. They master methods of interpretability and can optimise systems for interpretability. They are able to implement interpretability and explainability mechanisms for machine learning systems. |
| Contents: | The most significant disadvantage of machine learning and deep learning algorithms today: the interpretability of models. To trust predictions of real-life applications of AI it is important to understand how (Explainability) and why (Interpretability) a prediction is made.
- Key Concepts of Interpretability and Explainability Challenges
- Fundamentals of Feature Importance and Impact
- Global and Local Model-Agnostic Explainability Methods
- Anchor and Counterfactual Explanations
- Visualizing Convolutional Neural Networks
- Interpretation Methods for multivariate Forecasting and Sensitivity Analysis
- Tuning for Explainability
|
| Recommended Prerequisites: | Basic knowledge of programming and machine learning |
| Mandatory Prerequisites: | Knowledge of the content of module
- 11881: Foundations of Data Mining
or
- 12351: Grundlagen des Data Mining
|
| Forms of Teaching and Proportion: | -
Lecture
/ 2 Hours per Week per Semester
-
Laboratory training
/ 2 Hours per Week per Semester
-
Self organised studies
/ 120 Hours
|
| Teaching Materials and Literature: | - Script and presentations are available for download in Moodle at the beginning of the semester and on an ongoing basis. Problems for exercises and instructions for lab experiments can be downloaded.
- Serg Masis, Interpretable Machine Learning with Python: Learn to build interpretable high-performance models with hands-on real-world examples, Packt 2021
- Ajay Thampi, Interpretable Ai: Building Explainable Machine Learning Systems, Manning 2022
- Christoph Molnar, Interpretable Machine Learning: A Guide For Making Black Box Models Explainable, 2022
- Uday Kamath; John Liu, Explainable Artificial Intelligence: An Introductionto Interpretable Machine Learning, Springer 2021
|
| Module Examination: | Prerequisite + Final Module Examination (MAP) |
| Assessment Mode for Module Examination: | Prerequisite:- Successful completion of exercises and presentation of results in course
Final module examination:- Written examination, 120 min.
|
| Evaluation of Module Examination: | Performance Verification – graded |
| Limited Number of Participants: | None |
| Part of the Study Programme: | -
Master (research-oriented) /
Artificial Intelligence /
PO 2022
-
Master (research-oriented) /
Cyber Security /
PO 2017
-
Master (research-oriented) /
Informatik /
PO 2008
-
Master (research-oriented) /
Künstliche Intelligenz Technologie /
PO 2022
|
| Remarks: | - Study programme Informatik M.Sc.: Compulsory elective module in complex „Angewandte und Technische Informatik" (level 400)
- Study programme Artificial Intelligence M.Sc.: Compulsory elective module in complex „Learning and Reasoning“
- Study programme Künstliche Intelligenz Technologie M.Sc.: Compulsory elective module in complex „Software-basierte Systeme“
- Study programme Cyber Security M.Sc.: Compulsory elective module in complex „Computer Science"
|
| Module Components: | - Lecture: Explainable Machine Learning
- Accompanying laboratory
- Accompanying Examination
|
| Components to be offered in the Current Semester: | |
