PMLE Practice Test 25 — 15 Questions

Question 1

A data scientist needs to train a large PyTorch model on a custom dataset using Vertex AI. The training script expects data from Cloud Storage and uses GPU acceleration. Which option correctly configures a custom training job with a pre-built container for PyTorch and attaches a single NVIDIA V100 GPU?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Use the pre-built container 'us-docker.pkg.dev/vertex-ai/training/pytorch-gpu.1-12:latest' and in worker_pool_specs set machine_type='n1-standard-4', accelerator_type='NVIDIA_TESLA_V100', accelerator_count=1. In Vertex AI, worker_pool_specs define machine types and accelerators. Using a pre-built container for PyTorch 1.12 with image_uri 'us-docker.pkg.dev/vertex-ai/training/pytorch-gpu.1-12:latest' and specifying worker_count=1, machine_type='n1-standard-4' (or similar), and accelerator_type='NVIDIA_TESLA_V100' with count=1 sets up the job correctly.

Answer

Use a custom container built from PyTorch base image and specify accelerator_count=1 in the machine spec

Answer

Use the AI Platform Training service with gcloud ai-platform jobs submit training and --scale-tier BASIC_GPU

Answer

Create a training pipeline with AutoML and select GPU runtime

Question 2

A company wants to implement continuous training for their ML model. The pipeline should be triggered when new training data arrives in Cloud Storage, and after training, the model should be automatically deployed to a staging endpoint if evaluation metrics pass a threshold. They also need to detect skew between training data and serving data. Which two services should they use for skew detection?

Accepted Answer

Vertex AI Model Monitoring and Vertex AI Pipelines. Vertex AI Model Monitoring provides built-in skew detection by comparing training data statistics with serving data statistics, alerting when distribution shifts exceed thresholds. Vertex AI Pipelines orchestrates the continuous training workflow, including triggering on new data arrival, model evaluation, and conditional deployment to a staging endpoint, making C the correct pair for both skew detection and pipeline automation.

Answer

Cloud Monitoring and Cloud Logging

Answer

Cloud DLP and Cloud KMS

Answer

BigQuery and Dataflow

Question 3

An ML engineer needs to deploy a model to an endpoint and gradually shift traffic from the previous version (champion) to a new version (challenger) for A/B testing. How should they configure the endpoint?

Accepted Answer

Deploy both versions to the same endpoint and set traffic splitting. Vertex AI endpoints support traffic splitting by assigning percentages to different model versions.

Answer

Use a canary deployment with Cloud Run

Answer

Manually update the endpoint to point to the challenger after testing

Answer

Create a new endpoint for the challenger and route traffic via load balancer

Question 4

A company has a TensorFlow model that requires GPU for inference. They are deploying on Vertex AI. Which TWO configurations are necessary to ensure GPU is used?

Accepted Answer

Select a machine type that includes a GPU (e.g., NVIDIA Tesla T4).. Vertex AI requires you to explicitly select a machine type that includes a GPU (e.g., n1-standard-4 with an attached NVIDIA Tesla T4) to provide the physical hardware for GPU acceleration. Without selecting a GPU machine type, the inference will run on CPU only, regardless of any other configuration.

Answer

Set the environment variable TF_GPU_ALLOCATOR=cuda_malloc_async.

Answer

Use the pre-built TensorFlow serving container, which automatically uses GPU if available.

Answer

Build a custom container with GPU drivers.

Question 5

A company uses Vertex AI Feature Store with an online store for low-latency serving. They observe high latency during peak hours. The feature values are small (< 1 KB each) and the workload is read-heavy. Which change would most effectively reduce latency?

Accepted Answer

Switch from Bigtable online store to Optimized online store. Switching from Bigtable online store to Optimized online store is recommended for read-heavy workloads with small feature values, offering lower latency at high QPS.

Answer

Enable caching on the client side

Answer

Use a larger machine type for Bigtable

Answer

Increase the number of Bigtable nodes

Question 6

A data scientist is training a very large neural network using Vertex AI with multiple GPUs across multiple nodes. The model does not fit on a single GPU, so they need to use both data parallelism and model parallelism (pipeline parallelism). Which THREE components or configurations are required to set up distributed training with Vertex AI?

Accepted Answer

Implementing pipeline parallelism manually in the training script using torch.distributed.pipeline.sync.Pipe. Pipeline parallelism requires explicit implementation in the training script, such as using `torch.distributed.pipeline.sync.Pipe` in PyTorch, to split the model layers across multiple GPUs. This is necessary when the model does not fit on a single GPU, and Vertex AI does not automatically handle model parallelism—it must be coded by the user.

Answer

Using Vertex AI Vizier to optimize the model parallelism strategy

Answer

Enabling Vertex AI AutoML to automatically distribute the model

Question 7

A team is using Vertex AI Experiments to compare different hyperparameters. They want to automatically record the hyperparameters. What is the correct way?

Accepted Answer

Use the `aiplatform.start_run()` context manager. Vertex AI Experiments provides a native `aiplatform.start_run()` context manager that automatically captures hyperparameters passed as key-value arguments, logging them to the experiment run metadata without manual intervention. This integrates directly with the Vertex AI SDK, ensuring consistency and traceability across runs.

Answer

Manually log to console

Answer

Write to a CSV file

Answer

Use BigQuery

Question 8

A Vertex AI Pipeline contains a task that may produce outputs that are not always needed. The engineer wants to conditionally execute downstream tasks only if a specific artifact is produced. Which KFP SDK v2 construct allows the engineer to implement this conditional execution?

Accepted Answer

dsl.If. The dsl.If construct in KFP SDK v2 allows conditional execution of tasks based on the presence or value of an output from a previous task. dsl.Collected is for parallel loops, exit handlers are for cleanup, and dsl.Condition is not a valid construct.

Answer

dsl.ExitHandler

Answer

dsl.Collected

Answer

dsl.Condition

Question 9

A company wants to predict customer churn using a dataset with 10,000 rows and 20 features. They have no ML expertise. Which low-code solution should they use?

Accepted Answer

Vertex AI AutoML Tables. Vertex AI AutoML Tables is the correct low-code solution because it allows users with no ML expertise to train high-quality tabular models on structured data (10,000 rows, 20 features) without writing any code. It automates feature engineering, model selection, and hyperparameter tuning, and provides a simple UI to upload data and get predictions. This directly matches the requirement of a low-code, no-expertise solution for a tabular churn prediction problem.

Answer

Kubeflow Pipelines

Answer

Custom TensorFlow model

Answer

BigQuery ML

Question 10

A company has deployed a model for image classification and wants to monitor for feature drift using XRAI attributions. However, they notice that the XRAI attribution maps are too large and are causing high latency in the monitoring pipeline. What is the most effective way to reduce the overhead of explainability monitoring for image models?

Accepted Answer

Reduce the sampling rate for the explainability feature. Vertex AI Explainable AI supports XRAI for image models, but generating XRAI attributions can be computationally expensive. Sampling a subset of predictions reduces the number of explanations generated, lowering latency and cost.

Answer

Disable XRAI and use integrated gradients instead

Answer

Use a smaller image size for the model

Answer

Increase the number of replicas on the endpoint

Question 11

A data scientist wants to use a pre-trained ResNet model from Keras Applications and fine-tune it on a small custom dataset. Which approach should they take to avoid overfitting?

Accepted Answer

Freeze the first few layers and train the rest.. Freezing the earlier layers (which capture general features) and only training the later layers is a common transfer learning approach for small datasets, reducing overfitting.

Answer

Add more convolutional layers to the model.

Answer

Use a larger learning rate to speed up training.

Answer

Train the entire model from scratch on the custom dataset.

Question 12

You deploy a new version of a model to a Vertex AI endpoint and want to gradually shift traffic from the old version to the new version over 24 hours. The endpoint currently serves 100% traffic to the old version. What should you do?

Accepted Answer

Update the endpoint to split traffic between the two model versions using the traffic split configuration.. Vertex AI endpoints support a built-in traffic split configuration that allows you to gradually shift traffic between model versions deployed to the same endpoint. By updating the endpoint's traffic split percentages (e.g., from 100% old / 0% new to 0% old / 100% new over 24 hours), you can achieve a smooth, controlled rollout without changing client code or managing multiple endpoints.

Answer

Use Vertex AI Experiments to run an A/B test between the two versions.

Answer

Deploy the new version to a separate endpoint and update your client to use the new endpoint for a percentage of requests.

Answer

Delete the old version and redeploy the new version with a different endpoint name, then update DNS.

Question 13

A team uses Vertex AI Feature Store to serve features for real-time predictions. They notice that feature values are frequently updated from multiple source systems, leading to inconsistencies. They need to ensure that feature values are consistent across all serving endpoints. What should they do?

Accepted Answer

Use a streaming ingestion pipeline with exactly-once semantics. Streaming ingestion with exactly-once semantics ensures that each feature update is applied precisely once, preventing duplicates or missed updates that cause inconsistencies. This approach synchronizes feature values across all serving endpoints in near real-time, directly addressing the problem of frequent updates from multiple source systems.

Answer

Use batch ingestion with weekly updates to reduce update frequency

Answer

Increase the offline storage TTL to retain historical feature values

Answer

Implement a manual approval process for feature updates

Question 14

A machine learning engineer notices that the Vertex AI Prediction endpoint's error rate has increased over the past week. The model was retrained with new data and redeployed. Which step should the engineer take first to diagnose the issue?

Accepted Answer

Check the Cloud Monitoring dashboard for latency and error codes, and review the model's prediction logs.. The correct first step is D because checking Cloud Monitoring dashboards and reviewing prediction logs provides immediate visibility into error patterns and root causes, such as data shifts or infrastructure issues. Option A (increasing replicas) addresses symptoms but not the cause. Option B (Explainable AI) is more advanced and typically used after initial diagnostics. Option C (rolling back) may restore previous performance but does not identify why the new model failed, and could discard improvements.

Answer

Increase the number of replicas to reduce error rate.

Answer

Compare the input data distribution of recent requests to the training data distribution using Explainable AI.

Answer

Roll back to the previous model version immediately.

Question 15

An organization wants to use Cloud Composer (Airflow) to orchestrate a machine learning workflow that includes running a Vertex AI Pipeline, followed by a BigQuery job, and then a Dataflow pipeline. What is the primary advantage of using Cloud Composer for this orchestration?

Accepted Answer

It allows orchestrating heterogeneous workflows across multiple GCP services with dependencies and retries.. Cloud Composer (Apache Airflow) is designed to orchestrate heterogeneous workflows across multiple GCP services. In this scenario, it can define a Directed Acyclic Graph (DAG) that runs a Vertex AI Pipeline, then a BigQuery job, and finally a Dataflow pipeline, with built-in support for dependency management, retries, and failure handling. This is the primary advantage because it allows you to coordinate disparate services in a single, reliable workflow.

Answer

It automatically caches the outputs of each step to avoid recomputation.

Answer

It integrates natively with the Vertex AI Model Registry for model versioning.

Answer

It provides a serverless execution environment for ML pipelines.