Module 7: Responsible AI

Overview

Welcome to the Responsible AI module of the EDSP mentoring program!

What you’ll learn

• Microsoft’s Ethical Obligation & Responsible AI Principles
• Intepreting ML Model Behavior & Explaining Their Inferences
• Detecting & Mitigating Unintended Bias in Training Data and ML Models
• Microsoft Tools for Responsible AI: InterpretML, Fairlearn, and More

Topic Kickoff

Resources	Links
Presentation	Presentation
Recording	Recording

Table of Contents

Resources	Links
Recording	Model Explainability & Responsible AI with AML (60 minutes)
Recording	How to Explain Models with IntepretML Deep Dive - YouTube  (30 minutes)
Recording	How to Test Models for Fairness with Fairlearn Deep-Dive - YouTube  (12 minutes)
Recording	Explainable AI explained: Introduction  (10 mins)
Recording	Explainable AI explained: 2 By Design interpretable models with Microsoft InterpretML (20 mins)
Recording	Explainable AI explained: 3 LIME - YouTube (15 mins)
Recording	Explainable AI explained: 4 SHAP - YouTube (15 mins)
Tutorials: Jupyter Notebooks	Responsible AI Airlift (GitHub Repo)

Additional / Optional Resources

Resources	Links
Online Book	Interpretable Machine Learning, Christoph Molnar
Home Page	Microsoft AI - Responsible AI Resources
Code Repository	Responsible AI Toolbox
Software	InterpretML Library
Documentation	InterpretML Documentation
Software	Fairlearn Library
Documentation	Fairlearn Documentation