Google Exam Professional Machine Learning Engineer Topic 4 Question 84 Discussion

Actual exam question for Google's Professional Machine Learning Engineer exam

Question #: 84
Topic #: 4

[All Professional Machine Learning Engineer Questions]

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?

ACreate an AutoML tabular model by using the BigQuery data with integrated Vertex Explainable Al.

BCreate a BigQuery ML deep neural network model, and use the ML. EXPLAIN_PREDICT method with the num_integral_steps parameter.

CUpload the custom model to Vertex Al Model Registry and configure feature-based attribution by using sampled Shapley with input baselines.

DUpdate the custom serving container to include sampled Shapley-based explanations in the prediction outputs.

Show Suggested Answer

Suggested Answer: D

by Dolores at Aug 08, 2024, 02:58 AM

Limited Time Offer

25%

Off

Get Premium Professional Machine Learning Engineer Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Teddy

8 days ago

I'm not sure about option A. I think option C might provide more accurate explanations with feature-based attribution.

upvoted 0 times

...

Nakisha

10 days ago

I agree with Jacquelyne. Using AutoML with Vertex Explainable AI seems like the most efficient way to meet compliance requirements.

upvoted 0 times

...

Ronald

10 days ago

Option A seems like the easiest way to get explainability with minimal effort. AutoML Tabular models come with built-in Vertex Explainable AI, so that's an attractive choice.

upvoted 0 times

...

Jacquelyne

15 days ago

I think option A sounds like a good choice for adding explanations to the model.

upvoted 0 times

...