Administrative Information
Title | Guest Lecture on Explainable Machine Learning |
Duration | 60 |
Module | C |
Lesson Type | Lecture |
Focus | Technical - Future AI |
Topic | Open Problems and Challenges |
Keywords
Explainable AI,Interpretable models,Black Box models,Post-hoc,
Learning Goals
- Understand the importance of XAI
- Understand the difference between interpretable and explainable approaches
- Identify XAI stakeholders
- Understand the taxomony of XAI approaches
- Be able to indentifiy XAI limitations and the future direction of XAI
Expected Preparation
Learning Events to be Completed Before
Obligatory for Students
Optional for Students
References and background for students
- MIT 6.S191: Robust and Trustworthy Deep Learning - YouTube video
Recommended for Teachers
None.
Lesson materials
Instructions for Teachers
This guest lecture should explain the importance of explainable AI and the benefits it brings to AI systems. The lecture should identify the stakeholders involved and how XAI can be beneficial for each group (engineers, end users, legislators). A distinction should be drawn between interpretable AI and explainable AI along with a discussion of why XAI may not be good enough in the long run. The lecture should identify the desired characteristics of explainability and should outline a taxonomy of XAI approaches. The lecture should end with a discussion of the limitations of exisitng XAI approaches and their future directions.
Outline
Duration | Description |
---|---|
10 min | Importance of Explainable AI (XAI) |
10 min | Interpretable and Explainable AI |
10 min | XAI Stakeholders |
10 min | Taxonomy of XAI approaches |
10 min | XAI Limitiations and Future Direction |
Acknowledgements
The Human-Centered AI Masters programme was Co-Financed by the Connecting Europe Facility of the European Union Under Grant №CEF-TC-2020-1 Digital Skills 2020-EU-IA-0068.