How to explain Vertex AI forecast predictions?

You work for a retail company. You have created a Vertex AI forecast model that produces monthly item sales predictions. You want to quickly create a report that will help to explain how the model calculates the predictions. You have one month of recent actual sales data that was not included in the training dataset. How should you generate data for your report?

  1. Create a batch prediction job by using the actual sales data. Compare the predictions to the actuals in the report.
  2. Create a batch prediction job by using the actual sales data, and configure the job settings to generate feature attributions. Compare the results in the report. Source Reference Answer
  3. Generate counterfactual examples by using the actual sales data. Create a batch prediction job using the actual sales data and the counterfactual examples. Compare the results in the report.
  4. Train another model by using the same training dataset as the original, and exclude some columns. Using the actual sales data create one batch prediction job by using the new model and another one with the original model. Compare the two sets of predictions in the report.

Community Votes

B
100%

100% of anonymous learners picked answer B. Votes are pick records left by other test-takers — they are not the verified answer.

Community Insight

The question tests knowledge of Vertex AI Explainable AI, specifically requiring feature attributions to interpret model behavior rather than just evaluating accuracy.

To explain how a Vertex AI forecast model calculates predictions, you should configure a batch prediction job to generate feature attributions. This approach identifies the contribution of each input feature to the model's output.

Selecting Option A is a common error because comparing predictions to actual sales data measures model accuracy, not the reasoning behind the prediction.

Community Discussion (4 comments)

fitri001 👍 2 Selected: B
Feature Attribution: By enabling feature attributions in the batch prediction job, you gain insights into how each feature in the actual sales data contributes to the model's predictions. This information is crucial for explaining the model's reasoning to non-technical audiences. Direct Model Insights: Analyzing the feature attributions allows you to demonstrate how the model uses historical trends, seasonality, and other factors (represented by features) to predict future sales.
MultiCloudIronMan 👍 1 Selected: B
B is the best answer but am unsure why the report has to be compared with actual sales
ddogg 👍 3 Selected: B
B) Will actually give you the information needed with Feature Attributions. E.g. The importance of each feature influencing the predictions sales items.
pikachu007 👍 2 Selected: B
Feature attributions explicitly measure how much each input feature contributed to each prediction, providing the most relevant insights for understanding model behavior.

Comments & Corrections

No comments yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Submitted for moderation before publishing. Keep it helpful and respectful.

Expert Analysis

Why the Answer Is Correct

Option B is correct because Vertex AI provides Explainable AI features that allow you to generate feature attributions during a batch prediction job. These attributions quantify how much each input feature contributed to a specific prediction, directly satisfying the requirement to explain how the model calculates results.

Why the Other Options Are Wrong

Option A focuses on evaluating model performance by comparing predictions against ground truth data, which does not explain the model's internal logic. Option C involves counterfactual examples, which are useful for hypothetical scenarios but are not the standard method for explaining specific predictions. Option D is inefficient as it requires retraining a new model and relies on ablation rather than direct attribution methods.

Community Comment Notes

Community comments unanimously support Option B, noting that feature attributions provide the necessary insight into feature importance. Commenters emphasize that this data is crucial for demonstrating model reasoning to non-technical stakeholders and understanding how historical trends influence specific forecasts.

Official Reference

https://cloud.google.com/vertex-ai/docs/explainable-ai/overview

Exam Strategy

When a question asks to explain "how" a model makes a prediction, look for options involving "feature attributions" or "Explainable AI" rather than options focused on accuracy metrics or retraining.

Related Analysis

You work at a gaming startup that has several terabytes of s

A-Grade

You are an ML engineer at a retail company. You have built a

S-Grade

You are developing a recommendation engine for an online clo

S-Grade

Practice All PMLE Questions

Access 65 questions with complete answers and detailed explanations.

View Full PMLE Practice Test →

← Back to PMLE Study Guide