Professional Machine Learning Engineer

Question

You work for a telecommunications company and are creating a model to predict which customers might not pay their next phone bill. The model is intended to proactively offer at-risk customers support, such as service discounts and extensions to bill due dates. The data is stored in BigQuery, and the predictive features available for training the model are:

Customer_id
Age
Salary (measured in local currency)
Sex
Average bill value (measured in local currency)
Number of phone calls in the last month (integer)
Average duration of phone calls (measured in minutes)

You must investigate and mitigate potential bias against disadvantaged groups while maintaining model accuracy.

What should you do?

Explanation

Fairness should be assessed by comparing an explicitly defined performance metric, such as accuracy, across sensitive groups on a held-out test set. Sensitive attributes can be joined back to prediction results for this evaluation even when they are not used as training inputs. Excluding protected features or high-attribution features alone does not establish fairness, because other inputs can be proxy variables for protected characteristics. BigQuery ML supports evaluation of boosted-tree classifiers on evaluation data and reports classification accuracy.

Learn more

Professional Machine Learning Engineer

AstroTutor

Notes

Readiness

QuestionQ16

Community Discussion