Great job making it this far into the book! You now have various interpretability techniques in your toolkit, and you should be well equipped to build robust AI systems! This final part focuses on fairness and bias and paves the way for explainable AI.
In chapter 8, you will learn about various definitions of fairness and how you can check whether your model is biased. You’ll also learn about techniques to mitigate bias and a standardizing approach of documenting datasets using datasheets that help improve transparency and accountability with the stakeholders and users of the AI system.
In chapter 9, we’ll pave the way for explainable AI by discussing how to build such systems, and you’ll also learn about contrastive explanations using counterfactual examples.