Case context: A team of statisticians and machine learning engineers are working together on a project. A statistician is confused when an engineer says they need to reduce the 'Avoidable bias' of their model, as this doesn't match standard statistical terminology.

Question: As the project lead, how would you clarify the engineer's use of 'Avoidable bias' to the statistician to ensure team alignment? What is the technical equivalent of this term according to Andrew Ng?

Sample answer: I would explain that the engineer is using an informal definition optimized for practical algorithm improvement. To align with the statistician's vocabulary, I would clarify that 'Avoidable bias' in this context technically means 'error we attribute to the learning algorithm's bias that is over the optimal error rate.'

Key points:

• Acknowledge the terminological gap between statistics and ML practice
• Translate 'Avoidable bias' to 'error we attribute to the learning algorithm's bias that is over the optimal error rate'
• Emphasize the goal is practical algorithm improvement

Rubric: Full credit for correctly translating 'Avoidable bias' to its technical definition and explaining that the informal term is used for practical algorithm improvement.