PMLE Practice Test 19 — 15 Questions

Question 1

A company uses BigQuery to store feature data for ML training. A data engineer notices that a Vertex AI Training job is failing with 'Access Denied' errors when reading from a BigQuery table. The training job uses a custom service account that has been granted the 'bigquery.dataViewer' role on the dataset. What is the most likely cause of the failure?

Accepted Answer

The service account lacks the 'bigquery.jobs.create' permission in the project.. The 'bigquery.dataViewer' role grants permissions to read BigQuery data (e.g., bigquery.tables.getData), but it does not include the 'bigquery.jobs.create' permission. When a Vertex AI training job reads from BigQuery, it must first create a BigQuery job (a query job) to retrieve the data. Without 'bigquery.jobs.create' at the project level, the service account cannot initiate the read operation, resulting in an 'Access Denied' error even though it has data-level access.

Answer

The service account is not in the same project as the BigQuery dataset.

Answer

The BigQuery table is partitioned and requires row-level access.

Answer

The training job does not have the required network access to BigQuery.

Question 2

Which TWO tools can be used to collaborate on feature definitions across teams?

Accepted Answer

Vertex AI Feature Store. Options A and C are correct. Vertex AI Feature Store (A) allows sharing features, and Data Catalog (C) can catalog feature definitions. B (Cloud Storage) is not a collaboration tool. D (Cloud Build) is for CI/CD. E (Cloud Logging) is for logs.

Answer

Cloud Storage

Answer

Cloud Logging

Answer

Cloud Build

Question 3

An organization uses Cloud Dataflow to preprocess training data. Dataflow jobs are often failing because of insufficient quota for certain resources. The team has requested a quota increase, but the jobs still fail with 'quota exceeded' errors for a different resource. They want to proactively monitor and manage quotas to avoid failures. What is the best approach?

Accepted Answer

Set up Cloud Monitoring alerts for quota usage and automate quota increase requests.. Option A is correct because setting up Cloud Monitoring alerts for quota usage and automating quota increase requests helps catch issues before failures occur. Option B might reduce resource consumption but does not address the root cause of quota limits. Option C is not feasible. Option D could worsen the problem by requiring more resources.

Answer

Configure Dataflow to use a different pipeline type that avoids the quota.

Answer

Use Dataflow's autoscaling feature to reduce resource usage.

Answer

Increase the maximum number of workers in the Dataflow job.

Question 4

A company uses AutoML Tables to predict customer churn. The model's AUC is low. Which action is most likely to improve performance?

Accepted Answer

Add more training data. Adding more training data often helps improve model performance. Increasing the training budget alone may not help if data is insufficient. Removing features with low importance could hurt. Changing the optimization objective may not directly improve AUC.

Answer

Use a different optimization objective

Answer

Increase the training budget to 10 hours

Answer

Remove features with low importance

Question 5

A retailer uses BigQuery ML to build a linear regression model for sales forecasting. The model's evaluation shows high RMSE. Which step should they take first?

Accepted Answer

Examine the data for outliers and missing values. High RMSE in a linear regression model often indicates issues with data quality, such as outliers or missing values, which can disproportionately skew the model's predictions. BigQuery ML's linear regression is sensitive to such anomalies, so examining and cleaning the data is the most appropriate first step before considering model complexity or feature engineering.

Answer

Use a more complex model like XGBoost

Answer

Increase the number of features

Answer

Set a larger training budget

Question 6

A company deploys an AutoML Vision model for real-time defect detection. They notice high inference latency during peak hours. Which configuration change can help?

Accepted Answer

Increase the number of max replicas. Increasing the number of max replicas allows the AutoML Vision endpoint to scale horizontally during peak hours, distributing the inference load across more compute instances. This directly reduces per-request latency by preventing queuing and resource contention, as the Vertex AI Prediction service can spin up additional replicas up to the configured maximum to handle higher throughput.

Answer

Reduce the model's input resolution

Answer

Use batch prediction

Answer

Enable model compression

Question 7

A data scientist trained a custom TensorFlow model using Vertex AI Training and wants to deploy it for online predictions with low latency (<100ms). Which deployment option on Google Cloud is best?

Accepted Answer

Deploy on Vertex AI Endpoints. Vertex AI Endpoints is the correct choice because it is purpose-built for deploying TensorFlow models with optimized serving infrastructure, including automatic scaling, GPU/TPU support, and built-in monitoring for latency-sensitive online predictions. It provides a managed endpoint that can achieve sub-100ms latency by leveraging model optimization techniques like TensorFlow Serving and hardware accelerators, which are not available in the other options.

Answer

Deploy on Cloud Run with a custom container

Answer

Deploy on Cloud Functions

Answer

Deploy on AI Platform Prediction (legacy)

Question 8

An organization uses Vertex AI Pipelines to automate a model training workflow. They want to reuse previously trained models if the data hasn't changed. Which pipeline component best achieves this?

Accepted Answer

Use a conditional component that checks data hash. Conditional components can check data changes and skip training if unchanged. Caching exists but not for whole pipeline; Cloud Function check is external; Artifact Registry stores models but doesn't decide to retrain.

Answer

Use a caching mechanism in Vertex AI Pipelines

Answer

Use a Cloud Function to check BigQuery update time

Answer

Use Artifact Registry to store model versions

Question 9

A financial company is building a fraud detection model. The dataset has 1% fraud cases and 99% legitimate transactions. Which technique should they use to handle the class imbalance?

Accepted Answer

Use class weighting or synthetic oversampling (SMOTE) during training. Class weights or resampling techniques like SMOTE are standard for imbalanced datasets. Option A is correct. Option B (undersampling majority) can lose information. Option C (collect more data) is impractical. Option D (no alterations) will bias the model.

Answer

Randomly undersample the majority class to balance the dataset

Answer

Collect more data until the fraud rate increases

Answer

Train without any modifications; the model will naturally handle it

Question 10

A user receives this error when deploying an AutoML model. What should they do?

Accepted Answer

Use machine type n1-highmem-2. The error occurs because AutoML models require a machine type with sufficient memory to load the model and perform predictions. The default machine type may not have enough memory for the model's size, leading to an out-of-memory (OOM) error. Using `n1-highmem-2` provides higher memory per core, which resolves the memory constraint without changing other deployment parameters.

Answer

Change the region to us-west1

Answer

Increase the min-replica-count to 2

Answer

Remove the traffic-split flag

Question 11

A company needs to extract entities (e.g., names, dates) from customer emails using a pre-trained model. Which service should they use?

Accepted Answer

Natural Language API. The Natural Language API provides entity extraction as a pre-trained model. Vision API is for images, Translation API for translation, and Dialogflow for conversational agents.

Answer

Translation API

Answer

Dialogflow

Answer

Vision API

Question 12

A data scientist uses Vertex AI Pipelines to orchestrate an ML workflow. They want to reuse a component from Google's curated repository. What is the recommended way to incorporate it?

Accepted Answer

Use prebuilt components from the Google Cloud Pipeline Components repository. Option C is correct because Google provides a curated set of prebuilt components in the Google Cloud Pipeline Components repository, which are designed to be directly imported and used within Vertex AI Pipelines. These components encapsulate common ML tasks (e.g., model training, deployment) and are maintained by Google, ensuring compatibility and reducing custom code. Using them is the recommended approach to avoid reinventing the wheel and to leverage Google's best practices.

Answer

Import the component from Google Cloud Build

Answer

Use the 'aiplatform' Python SDK to define the component

Answer

Copy the component code into the pipeline definition

Question 13

A financial institution wants to use Natural Language API for sentiment analysis on customer feedback, but the domain-specific language (e.g., 'bullish', 'bearish') is not correctly classified. They have 200 labeled examples. Which approach minimizes coding effort while improving accuracy?

Accepted Answer

Use AutoML Natural Language to train a custom model. Option D is correct because AutoML Natural Language enables you to train a custom model on your 200 labeled examples without writing code, directly improving accuracy for domain-specific terms like 'bullish' and 'bearish'. This approach leverages transfer learning from Google's pre-trained models, minimizing coding effort while adapting to your unique vocabulary and sentiment patterns.

Answer

Submit a feature request to Google for domain-specific terms

Answer

Create a custom sentiment dictionary and pass it to the Natural Language API

Answer

Build a custom TensorFlow model for sentiment

Question 14

An e-commerce company deployed a Vertex AI AutoML Tables model to predict customer churn. The model is served via a private endpoint with a dedicated machine type n1-standard-4. After a week, they observe that 5% of predictions fail with 'Request timed out' error. The average prediction time is 1.2 seconds but spikes to 4 seconds during peak hours. The input data is 50 features. They have enabled autoscaling with a min node count of 1 and max of 5. Which action is most likely to resolve the timeout issue without increasing complexity?

Accepted Answer

Increase the max node count to 10.. Option A is correct because increasing the max node count allows the endpoint to handle peak traffic better. Option B (increasing memory) does not address compute demand. Option C (reducing features) changes the model and may degrade performance. Option D (model monitoring) does not fix the timeout.

Answer

Reduce the number of features to 30.

Answer

Enable model monitoring to detect data drift.

Answer

Change the machine type to n1-highmem-4 to increase memory.

Question 15

What is the root cause of the failure?

Accepted Answer

The budget_milli_node_hours parameter is set to 0, which is below the minimum required value. Option A is correct because the `budget_milli_node_hours` parameter in Vertex AI AutoML training specifies the maximum amount of compute time (in milliseconds) allocated for model training. Setting it to 0 means no compute time is allowed, which causes the training job to fail immediately as it cannot proceed with zero resource allocation. The minimum required value is typically 1 (or higher depending on the task type), so 0 is invalid and triggers a failure.

Answer

The evaluate_model component expects the model artifact but the autopilot_train component does not output a model artifact

Answer

The location parameter 'us-central1' is not a valid region for AutoML

Answer

The threshold parameter is missing in the autopilot_train component