You need to build an ML model for a social media application to predict whether a user's submitted profile photo meets the requirements. The application will inform the user if the picture meets the…

Question

Options

• AUse AutoML to optimize the model's recall in order to minimize false negatives.
• BUse AutoML to optimize the model's F1 score in order to balance the accuracy of false positives
• CUse Vertex AI Workbench user-managed notebooks to build a custom model that has three times
• DUse Vertex AI Workbench user-managed notebooks to build a custom model that has three times

• A
  10% (4)
• B
  70% (28)
• C
  5% (2)
• D
  15% (6)

Explanation

The goal is to not falsely accept a non-compliant photo, which means minimizing false positives (predicting 'compliant' when the photo is actually non-compliant). Optimizing for recall alone (A) minimizes false negatives but may increase false positives - the opposite of what's needed. The F1 score (B) is the harmonic mean of precision and recall, balancing both types of errors. This is the right trade-off: you want to reject non-compliant photos accurately (high precision) without being so aggressive that compliant photos are frequently rejected (maintained recall). Building a custom model with three times more data (C/D) is over-engineering when AutoML can be tuned with the right objective.

No community discussion yet for this question.