Microsoft Reactor

The XAI course provides a comprehensive overview of explainable AI, covering both theory and practice, and exploring various use cases for explainability. Participants will learn not only how to generate explanations, but also how to evaluate and effectively communicate these explanations to diverse stakeholders.

The XAI course is managed on a voluntary basis by DataNights and Microsoft organizers and free for charge for the participant. This course is designed for data scientists that have at least two years in industry of hands-on experience with machine learning and Python and a basic background in deep learning. Some of the sessions will be held in-person at the Microsoft Reactor in Tel Aviv, while others will be conducted virtually.

Course Leaders:
Bitya Neuhof, DataNights
Yasmin Bokobza, Microsoft

What is this session about?
In this introduction lecture on explainability in AI, we will delve into the key topics that surround this emerging field. We will first provide an overview of the motivation for explainability, exploring how it helps us to achieve more transparent and trustworthy AI systems, particularly from a managerial perspective. We will then define some of the key terminology in the field and differentiate between black box explanation and interpretable ML. We will discuss the differences between global and local explanations, and include many examples from different fields and use cases throughout the lecture.

Next, we will examine the "built-in" feature importance methods that are commonly used for regression and trees, and discuss the strengths and limitations of these methods.
Overall, this lecture will provide a comprehensive introduction to explainability in AI, covering the key topics and terminology that are essential for understanding this field.