Skip to content
ML Academy

Classical ML

Interview question

  • stafflevel
  • 25 minsuggested time
  • 2Data scientist, ML engineer

A lender's credit-decision model approves 30,000 of 100,000 applicants (hypothetical) and observes repayment only for approved loans, where 4.0% (1,200) default. The team wants to retrain the model next quarter on the 'known outcomes' and says "we have 30,000 labelled examples and good AUC on them". As the staff data scientist, what are the risks and what do you propose?

Tests: Selective labels and reject inference, feedback loops between decisions and training data, honest evaluation when outcomes are missing for rejected cases, and the economics of exploration.

day 2day 15day 19

Sign in to practise: answers are saved before model answers are shown.