Google Cloud Professional Machine Learning Engineer — Question 184

You work for a bank. You have created a custom model to predict whether a loan application should be flagged for human review. The input features are stored in a BigQuery table. The model is performing well, and you plan to deploy it to production. Due to compliance requirements the model must provide explanations for each prediction. You want to add this functionality to your model code with minimal effort and provide explanations that are as accurate as possible. What should you do?

Answer options

• A. Create an AutoML tabular model by using the BigQuery data with integrated Vertex Explainable AI.
• B. Create a BigQuery ML deep neural network model and use the ML.EXPLAIN_PREDICT method with the num_integral_steps parameter.
• C. Upload the custom model to Vertex AI Model Registry and configure feature-based attribution by using sampled Shapley with input baselines.
• D. Update the custom serving container to include sampled Shapley-based explanations in the prediction outputs.

Correct answer: C

Explanation

The correct answer is C because uploading the model to Vertex AI Model Registry allows for the integration of advanced explanation methods like feature-based attribution using Shapley values, which is suitable for compliance requirements. Option A does not allow for the same level of customization needed for a custom model. Option B is not the best choice as it requires using a specific deep learning approach rather than leveraging existing capabilities. Option D, while it could work, requires more manual effort and customization compared to using Vertex AI's built-in features.