Explainable AI: Two Key Techniques (Pilot)

Type of course:

Digital learning, Lesson

Language:

Duration:

15 minutes

Workload:

Array hour

Proficiency:

Intermediate

Target:

Professionals, Manager, Students

SUMMARY

This course introduces two powerful methods for explainable AI: Local Interpretable Model-agnostic Explanations (LIME) and Before and After Correction Parameter Comparison (BAPC). As AI systems increasingly impact critical decision-making processes, understanding how these models arrive at their predictions is essential for ensuring transparency, trust, and fairness. Through this course, participants will explore how LIME and BAPC, both model-agnostic and locally focused approaches, provide insights into the behavior of complex machine learning models.

About The Author

Software Competence Center Hagenberg (SCCH), is a research organization committed to building the bridge between fundamental research and industry. With more than 20 years of experience, SCCH has been bringing the highest value to academic research by implementing feasible solutions to companies around Europe in light of Industry 5.0 and the rise of artificial intelligence. We are composed of a great team of data and software scientists from Austria and many other nationalities, with great experience not only in technical development and implementation but also highly motivated to train our customers to use our solutions.

Learning outcomes

By the end of this lesson, students will be able to describe the reasoning behind AI explainability
By the end of this lesson, the learner will be able to explain how to compute feature importance to explain the prediction of an instancy by an AI model
By the end of this lesson, students will be able to distinguish between two explainable AI methods and highlight their advantages