Tutorial 11: Explainable AI (XAI)

This week we discuss explainability, a field of AI ethics which looks at increasing human understanding of the decisions made by AI systems. Explanations answer the question of why our models produce the results they do.

The aims of the tutorial on Explainability are as follows:

1. To critically assess the important factors that we want in explainability

2. To evaluate different explainability approaches on a real example.