Question: Analyze the difference between the informal definitions of 'Bias' and 'Avoidable bias' used by Andrew Ng in Machine Learning Yearning and their formal statistical counterparts. Why does he choose to use these informal definitions instead of the technical ones?

Sample answer: Andrew Ng uses informal definitions of bias to provide practical insights on how to improve learning algorithms, rather than focusing on theoretical purity. Formally, what he calls 'Bias' is the 'Error we attribute to bias,' and 'Avoidable bias' is the 'error we attribute to the learning algorithm's bias that is over the optimal error rate.' The informal terms are more direct and actionable for a machine learning practitioner.

Key points:

• Pedagogical focus on practical algorithm improvement
• Formal 'Bias' is 'Error we attribute to bias'
• Formal 'Avoidable bias' relates to error over the optimal error rate

Rubric: Full credit for stating the formal definition of Bias, the formal definition of Avoidable Bias, and the pedagogical reason for using informal definitions.