PMLE Practice Test 48 — 15 Questions

Question 1

An ML team wants to monitor feature drift in their production model. Which Vertex AI Feature Store capability should they use?

Accepted Answer

Feature monitoring (drift detection). Vertex AI Feature Store includes feature monitoring that can detect drift between training and serving data.

Answer

Feature views

Answer

Online store

Answer

Point-in-time retrieval

Question 2

You want to use a pre-trained model from TensorFlow Hub for image classification, but you need to adapt it to classify your own custom categories with a small dataset. Which Vertex AI approach is most appropriate?

Accepted Answer

Write a custom training script that loads the pre-trained model and fine-tunes it on your dataset. Transfer learning fine-tunes a pre-trained model on a new dataset with small data. JumpStart deploys foundation models but not fine-tune for custom categories easily. Custom container is overkill. AutoML requires no code but may not be suitable if you want to control the pre-trained model.

Answer

Deploy the pre-trained model as-is via Vertex AI JumpStart

Answer

Build a custom container with the pre-trained model and deploy to Vertex AI Endpoints

Answer

Use Vertex AI AutoML for image classification

Question 3

An ML engineer wants to use Vertex AI Model Garden to deploy a pre-trained foundation model for text summarisation. What is the quickest way to achieve this?

Accepted Answer

Use Vertex AI JumpStart to deploy the model with one click. Vertex AI JumpStart provides one-click deployment of foundation models like Llama, Gemini, etc. Model Garden is for discovery, but JumpStart directly deploys.

Answer

Use Vertex AI AutoML for text summarisation

Answer

Use Vertex AI Pipelines to build a custom training pipeline

Answer

Export the model from Model Garden and deploy using a custom container

Question 4

A company is deploying a model on Vertex AI for online predictions with strict latency SLOs. The model requires GPU acceleration. Which TWO configurations should they consider to meet the SLOs while optimizing cost?

Accepted Answer

Set min_replica_count to handle base traffic and max_replica_count to handle spikes. Enabling autoscaling with min_replica_count to handle base load and max_replica_count for spikes, and using scale-to-zero for non-production (but for production, scale-to-zero may not meet SLOs due to cold starts; however, the question says 'optimizing cost', so scale-to-zero is not appropriate for low latency. The correct answers are: set appropriate min and max replicas, and use GPU-enabled machine types. The other options are either irrelevant or counterproductive.

Answer

Use n1-highmem-32 machine types without GPU

Answer

Enable autoscaling with min_replica_count=0 and max_replica_count=10

Answer

Disable autoscaling and set a fixed number of replicas equal to peak load

Question 5

Which Vertex AI service is designed for building and managing approximate nearest neighbor (ANN) indexes for similarity search at scale?

Accepted Answer

Vertex AI Matching Engine (Vector Search). Vertex AI Matching Engine (now Vector Search) provides ANN indexes for similarity search, enabling fast vector similarity queries at scale.

Answer

Vertex AI AutoML

Answer

Vertex AI Workbench

Answer

Vertex AI Prediction

Question 6

You are tasked with building a robust ML pipeline that must be idempotent and handle data skew between training and serving. Which three practices should you implement?

Accepted Answer

Store intermediate data in Cloud Storage with unique run IDs.. Idempotent components ensure the same inputs produce the same outputs. Passing data via GCS URIs is a best practice. Skew detection should compare training data distribution with serving data. Using unique run IDs for outputs ensures idempotency. Avoiding in-memory data passing is important for large datasets.

Answer

Pass large datasets between components as serialized in-memory objects.

Answer

Use the same random seed for every run to ensure reproducibility.

Question 7

An ML engineer needs to monitor the online prediction latency of a Vertex AI Endpoint. Which metrics should they look at in Cloud Monitoring?

Accepted Answer

p50, p95, p99 latency. Cloud Monitoring provides latency metrics for Vertex AI Endpoints, including p50, p95, and p99 latency, which are key for understanding performance.

Answer

Request count and error rate

Answer

Skew and drift scores

Answer

CPU/GPU utilization

Question 8

An MLOps team has deployed a model on Vertex AI Endpoints and wants to monitor for skew between training and serving data distributions. Which Vertex AI service should they use?

Accepted Answer

Vertex AI Model Monitoring. Vertex AI Model Monitoring is specifically designed for monitoring feature skew (training vs serving) and drift (serving over time) on deployed models.

Answer

Vertex AI Explainability

Answer

Vertex AI Model Registry

Answer

Vertex AI Continuous Training

Question 9

Which algorithm does Vertex AI Model Monitoring use by default to detect feature drift in a categorical feature?

Accepted Answer

Jensen-Shannon divergence. Vertex AI Model Monitoring uses Jensen-Shannon divergence (JS divergence) as the default metric for drift detection on categorical features.

Answer

Population Stability Index (PSI)

Answer

Wasserstein distance

Answer

Kullback-Leibler divergence

Question 10

A company deploys a model on Vertex AI Endpoint and expects high traffic spikes during promotional events. The current configuration uses manual scaling with 2 replicas. Which autoscaling configuration should they use to handle spikes while minimizing cost during normal traffic?

Accepted Answer

Enable basic scaling with target_cpu_utilization=0.6 and set min_replica_count=2, max_replica_count=10.. Basic scaling with a target CPU utilization (e.g., 0.6) automatically adjusts the number of replicas between min and max based on actual load. This allows the endpoint to scale up during traffic spikes (up to 10 replicas) to handle high demand, and scale down to the minimum (2 replicas) during normal traffic to minimize cost. Option A is wrong because manual scaling with fixed 10 replicas would incur unnecessary cost during low traffic and still may not handle extreme spikes if load exceeds 10 replicas. Option B is wrong because without a scaling metric, the endpoint cannot determine when to scale; it needs a metric like CPU utilization or request count. Option D is wrong because while custom metric scaling is possible, basic scaling with CPU utilization is simpler, sufficient, and recommended for CPU-bound models; the question asks for the configuration to handle spikes while minimizing cost, and basic scaling with CPU target achieves that.

Answer

Keep manual scaling but increase replicas to 10.

Answer

Set min_replica_count=2 and max_replica_count=10 with no scaling metric.

Answer

Use custom metric scaling with a Cloud Monitoring metric for prediction latency.

Question 11

You are an ML engineer at a fintech company. You have a prototype credit risk model built using XGBoost that achieves high accuracy on historical data. The model is trained on a dataset with 500,000 rows and 50 features. The company wants to deploy this model to production to score loan applications in real-time. The production environment must handle a peak load of 100 requests per second with a latency under 200ms. You have decided to use Vertex AI for deployment. After deploying the model as a Vertex AI endpoint with a single n1-standard-4 machine, you notice that latency exceeds 500ms at peak load and some requests time out. You have verified that the model prediction itself (excluding network overhead) takes about 50ms on average. What should you do to meet the latency and throughput requirements?

Accepted Answer

Enable autoscaling with a minimum of 2 replicas and use a larger machine type (e.g., n1-standard-8) to handle more concurrent requests.. The latency bottleneck is not the model inference time (50ms) but the inability of a single n1-standard-4 machine to handle 100 concurrent requests per second without queuing. By enabling autoscaling with a minimum of 2 replicas and upgrading to n1-standard-8, you increase both the number of concurrent requests the endpoint can process and the CPU/memory resources per replica, reducing queue wait times and keeping total latency under 200ms. This directly addresses the throughput and latency requirements without changing the model or switching to batch processing.

Answer

Change the machine type to a GPU-accelerated machine like n1-standard-4 with a T4 GPU.

Answer

Prune the model to reduce size and improve prediction speed.

Answer

Switch from online prediction to batch prediction using Vertex AI Batch Prediction.

Question 12

A team of data scientists and ML engineers is collaborating on a project using Vertex AI Workbench. They need to share notebooks and code, but want to avoid conflicts and maintain a history of changes. Which approach should they use?

Accepted Answer

Use a git repository (e.g., Cloud Source Repositories) to manage code and notebooks.. Using a git repository (e.g., Cloud Source Repositories) provides version control, branching, and a full history of changes, which is essential for collaborative development. This approach avoids conflicts by allowing team members to work on separate branches and merge changes systematically, unlike shared storage or manual methods that lack conflict resolution and audit trails.

Answer

Email notebook files to each other and manually merge changes.

Answer

Store notebooks in a shared Cloud Storage bucket and access them simultaneously.

Answer

Use Vertex AI Experiments to share notebook outputs.

Question 13

A data analyst wants to create a classification model directly in BigQuery using SQL. Which feature should they use?

Accepted Answer

BigQuery ML. BigQuery ML (BQML) enables users to create and execute machine learning models directly in BigQuery using standard SQL syntax, without needing to export data or manage separate ML infrastructure. For a data analyst who wants to build a classification model entirely within BigQuery, BQML provides the CREATE MODEL statement with classification algorithms like logistic regression or XGBoost, making it the correct and most direct feature.

Answer

Vertex AI

Answer

Dataflow

Answer

Cloud ML Engine

Question 14

A company wants to run batch predictions on millions of records stored in BigQuery. They need to preprocess the data (e.g., feature engineering) before feeding it to the model. Which approach is most scalable and cost-effective?

Accepted Answer

Preprocess with Cloud Dataflow, output to Cloud Storage, then submit a Vertex AI batch prediction job.. The most scalable and cost-effective because Cloud Dataflow (Apache Beam) provides serverless, auto-scaling preprocessing that handles large volumes of data efficiently, and Vertex AI batch predictions natively read from Cloud Storage, avoiding the need to manage infrastructure. This decouples preprocessing from prediction, allowing each to scale independently and minimizing costs by using ephemeral, pay-per-use resources.

Answer

Use a large DataProc cluster to preprocess and run batch predictions.

Answer

Preprocess inline in the batch prediction job using a custom container.

Answer

Use a custom Python script on a Compute Engine instance.

Question 15

Refer to the exhibit. A machine learning engineer deployed a model on Vertex AI using this configuration. When testing the endpoint, the engineer receives a 400 error with the message: 'Invalid argument: Explanation metadata missing required field: `outputs`.' What is the most likely cause?

Accepted Answer

The explanation metadata outputs field is missing the required 'displayName' attribute.. The error message 'Invalid argument: Explanation metadata missing required field: `outputs`' indicates that the `outputs` field is entirely absent from the explanation metadata configuration. None of the provided options correctly identifies this issue. Option A incorrectly attributes the error to a missing 'displayName' attribute within an existing outputs field, which is not the cause.

Answer

The explanation metadata needs a 'baseline' configuration for the input.

Answer

The explanation metadata inputs field should be wrapped inside a 'visualization' block.

Answer

The explainability method chosen is not supported for the model type.