Google Professional Machine Learning Engineer

• #121

  You are employed at a subscription-based company. You have trained an ensemble of tree and neural network models to forecast customer churn, which is the probability that customers will not renew their annual subscriptions. While the average prediction indicates a 15% churn rate, a specific customer is forecasted to have a 70% likelihood of churning. This customer has a product usage history of 30%, resides in New York City, and has been a customer since 1997. Your objective is to elucidate the distinction between the individual prediction of a 70% churn rate and the average prediction. To achieve this, you intend to employ Vertex Explainable AI. What is your recommended course of action?

  • AA. Train local surrogate models to explain individual predictions.
  • BB. Configure sampled Shapley explanations on Vertex Explainable AI.
  • CC. Configure integrated gradient explanations on Vertex Explainable AI.
  • DD. Measure the effect of each feature as the weight of the feature multiplied by the feature value.
  Show answer
• #122

  You have developed and are managing a production system tasked with predicting sales figures. The accuracy of this model is of paramount importance, as it needs to adapt to market fluctuations. Although the model has remained unchanged since its deployment, there has been a consistent decline in its accuracy. What could be the primary reason for this gradual decrease in model accuracy?

  • AA. Poor data quality
  • BB. Lack of model retraining
  • CC. Too few layers in the model for capturing information
  • DD. Incorrect data split ratio during model training, evaluation, validation, and test
  Show answer
• #123

  You have a model that is trained using data from a third-party data broker, and you are facing challenges because the data broker does not consistently inform you about formatting changes in the data. You aim to enhance the resilience of your model training pipeline to address such issues. What steps should you take?

  • AA. Use TensorFlow Data Validation to detect and flag schema anomalies.
  • BB. Use TensorFlow Transform to create a preprocessing component that will normalize data to the expected distribution, and replace values that don’t match the schema with 0.
  • CC. Use tf.math to analyze the data, compute summary statistics, and flag statistical anomalies.
  • DD. Use custom TensorFlow functions at the start of your model training to detect and flag known formatting errors.
  Show answer
• #124

  You work at an ecommerce startup. You need to create a customer churn prediction model. Your company’s recent sales records are stored in a BigQuery table. You want to understand how your initial model is making predictions. You also want to iterate on the model as quickly as possible while minimizing cost. How should you build your first model?

  • AA. Export the data to a Cloud Storage bucket. Load the data into a pandas DataFrame on Vertex AI Workbench and train a logistic regression model with scikit-learn.
  • BB. Create a tf.data.Dataset by using the TensorFlow BigQueryClient. Implement a deep neural network in TensorFlow.
  • CC. Prepare the data in BigQuery and associate the data with a Vertex AI dataset. Create an AutoMLTabularTrainingJob to tram a classification model.
  • DD. Export the data to a Cloud Storage bucket. Create a tf.data.Dataset to read the data from Cloud Storage. Implement a deep neural network in TensorFlow.
  Show answer
• #125

  You are developing an ML model using a dataset with categorical input variables. You have randomly split half of the data into training and test sets. After applying one-hot encoding on the categorical variables in the training set, you discover that one categorical variable is missing from the test set. What should you do?

  • AA. Use sparse representation in the test set.
  • BB. Randomly redistribute the data, with 70% for the training set and 30% for the test set.
  • CC. Apply one-hot encoding on the categorical variables in the test data.
  • DD. Collect more data representing all categories.
  Show answer
• #126

  You've trained a text classification model in TensorFlow on AI Platform and now need to perform batch predictions on text data stored in BigQuery, all while minimizing computational overhead. What's the recommended approach?

  • AA. Export the model to BigQuery ML.
  • BB. Deploy and version the model on AI Platform.
  • CC. Use Dataflow with the SavedModel to read the data from BigQuery.
  • DD. Submit a batch prediction job on AI Platform that points to the model location in Cloud Storage.
  Show answer
• #127

  You work as an ML engineer at an ecommerce company, and your current assignment involves constructing a model for forecasting the optimal monthly inventory orders for the logistics team. How should you proceed with this task?

  • AA. Use a clustering algorithm to group popular items together. Give the list to the logistics team so they can increase inventory of the popular items.
  • BB. Use a regression model to predict how much additional inventory should be purchased each month. Give the results to the logistics team at the beginning of the month so they can increase inventory by the amount predicted by the model.
  • CC. Use a time series forecasting model to predict each item's monthly sales. Give the results to the logistics team so they can base inventory on the amount predicted by the model.
  • DD. Use a classification model to classify inventory levels as UNDER_STOCKED, OVER_STOCKED, and CORRECTLY_STOCKED. Give the report to the logistics team each month so they can fine-tune inventory levels.
  Show answer
• #128

  You have recently created a deep learning model using Keras and are currently exploring various training strategies. Initially, you trained the model on a single GPU, but the training process proved to be too slow. Subsequently, you attempted to distribute the training across 4 GPUs using tf.distribute.MirroredStrategy, but you did not observe a reduction in training time. What steps should you take next?

  • AA. Distribute the dataset with tf.distribute.Strategy.experimental_distribute_dataset
  • BB. Create a custom training loop.
  • CC. Use a TPU with tf.distribute.TPUStrategy.
  • DD. Increase the batch size.
  Show answer
• #129

  You're developing a serverless ML system architecture to augment customer support tickets with relevant metadata before routing them to support agents. The system requires models for predicting ticket priority, estimating resolution time, and conducting sentiment analysis, aiding agents in strategic decision-making during support request processing. The tickets are anticipated to be free of domain-specific terminology or jargon. The proposed architecture will follow this sequence: Which endpoints should the Enrichment Cloud Functions call?

  • AA. 1 = AI Platform, 2 = AI Platform, 3 = Vertex AI AutoML Vision
  • BB. 1 = AI Platform, 2 = AI Platform, 3 = Vertex AI AutoML Natural Language
  • CC. 1 = AI Platform, 2 = AI Platform, 3 = Cloud Natural Language API
  • DD. 1 = Cloud Natural Language API, 2 = AI Platform, 3 = Cloud Vision API
  Show answer
• #130

  You are working on a prototype of a text classification model in a managed Vertex AI Workbench notebook. You want to quickly experiment with tokenizing text by using a Natural Language Toolkit (NLTK) library. How should you add the library to your Jupyter kernel?

  • AA. Install the NLTK library from a terminal by using the pip install nltk command.
  • BB. Write a custom Dataflow job that uses NLTK to tokenize your text and saves the output to Cloud Storage.
  • CC. Create a new Vertex AI Workbench notebook with a custom image that includes the NLTK library.
  • DD. Install the NLTK library from a Jupyter cell by using the !pip install nltk --user command.
  Show answer
• #131

  Your company manages an ecommerce website. You developed an ML model that recommends additional products to users in near real time based on items currently in the user’s cart. The workflow will include the following processes: - The website will send a Pub/Sub message with the relevant data and then receive a message with the prediction from Pub/Sub - Predictions will be stored in BigQuery - The model will be stored in a Cloud Storage bucket and will be updated frequently You want to minimize prediction latency and the effort required to update the model. How should you reconfigure the architecture?

  • AA. Write a Cloud Function that loads the model into memory for prediction. Configure the function to be triggered when messages are sent to Pub/Sub.
  • BB. Create a pipeline in Vertex AI Pipelines that performs preprocessing, prediction, and postprocessing. Configure the pipeline to be triggered by a Cloud Function when messages are sent to Pub/Sub.
  • CC. Expose the model as a Vertex AI endpoint. Write a custom DoFn in a Dataflow job that calls the endpoint for prediction.
  • DD. Use the RunInference API with WatchFilePattern in a Dataflow job that wraps around the model and serves predictions.
  Show answer
• #132

  You are training a custom language model for your company using a large dataset, and you plan to use the Reduction Server strategy on Vertex AI. You need to configure the worker pools for the distributed training job. What should you do?

  • AA. Configure the machines of the first two worker pools to have GPUs and use a container image where your training code runs. Configure the third worker pool to have GPUs and use the reductionserver container image.
  • BB. Configure the machines of the first two worker pools to have GPUs and use a container image where your training code runs. Configure the third worker pool without accelerators, use the reductionserver container image, and choose a machine type that prioritizes bandwidth.
  • CC. Configure the machines of the first two worker pools to have TPUs and use a container image where your training code runs. Configure the third worker pool without accelerators, use the reductionserver container image, and choose a machine type that prioritizes bandwidth.
  • DD. Configure the machines of the first two worker pools to have TPUs and use a container image where your training code runs. Configure the third worker pool to have TPUs and use the reductionserver container image.
  Show answer
• #133

  You work for a large bank with an application hosted on Google Cloud, operating in the US and Singapore. You've developed a PyTorch model, a three-layer perceptron, to classify transactions as potentially fraudulent. This model uses both numerical and categorical features, with hashing conducted within the model itself. Currently, the model is deployed in the us-central1 region on nl-highcpu-16 machines, delivering predictions with a median response latency of 40 ms. To reduce latency, particularly for users in Singapore who are experiencing longer delays, what should you do?

  • AA. Attach an NVIDIA T4 GPU to the machines currently used for online inference.
  • BB. Switch the machines used for online inference to nl-highcpu-32.
  • CC. Deploy the model to Vertex AI private endpoints in both the us-central1 and asia-southeast1 regions, allowing the application to select the most suitable endpoint.
  • DD. Establish another Vertex AI endpoint in the asia-southeast1 region, enabling the application to select the most appropriate endpoint.
  Show answer
• #134

  You are using Vertex AI and TensorFlow to develop a custom image classification model. You need the model’s decisions and the rationale to be understandable to your company’s stakeholders. You also want to explore the results to identify any issues or potential biases. What should you do?

  • AA. 1. Use TensorFlow to generate and visualize features and statistics. 2.Analyze the results together with the standard model evaluation metrics.
  • BB. 1. Use TensorFlow Profiler to visualize the model execution. 2.Analyze the relationship between incorrect predictions and execution bottlenecks.
  • CC. 1. Use Vertex Explainable AI to generate example-based explanations. 2.Visualize the results of sample inputs from the entire dataset together with the standard model evaluation metrics.
  • DD. 1. Use Vertex Explainable AI to generate feature attributions. Aggregate feature attributions over the entire dataset. 2.Analyze the aggregation result together with the standard model evaluation metrics.
  Show answer
• #135

  You have received a training-serving skew alert from a Vertex AI Model Monitoring job that is active in a production environment. In response, you have retrained the model using more up-to-date training data and subsequently redeployed it to the Vertex AI endpoint. Despite these actions, you continue to receive the same alert. What step should you take to address this situation?

  • AA. Update the model monitoring job to use a lower sampling rate.
  • BB. Update the model monitoring job to use the more recent training data that was used to retrain the model.
  • CC. Temporarily disable the alert. Enable the alert again after a sufficient amount of new production traffic has passed through the Vertex AI endpoint.
  • DD. Temporarily disable the alert until the model can be retrained again on newer training data. Retrain the model again after a sufficient amount of new production traffic has passed through the Vertex AI endpoint.
  Show answer
• #136

  You work for a retailer that sells clothes to customers around the world. You have been tasked with ensuring that ML models are built in a secure manner. Specifically, you need to protect sensitive customer data that might be used in the models. You have identified four fields containing sensitive data that are being used by your data science team: AGE, IS_EXISTING_CUSTOMER, LATITUDE_LONGITUDE, and SHIRT_SIZE. What should you do with the data before it is made available to the data science team for training purposes?

  • AA. Tokenize all of the fields using hashed dummy values to replace the real values.
  • BB. Use principal component analysis (PCA) to reduce the four sensitive fields to one PCA vector.
  • CC. Coarsen the data by putting AGE into quantiles and rounding LATITUDE_LONGTTUDE into single precision. The other two fields are already as coarse as possible.
  • DD. Remove all sensitive data fields, and ask the data science team to build their models using non-sensitive data.
  Show answer
• #137

  As a data scientist at an industrial equipment manufacturing company, you are currently working on creating a regression model. This model aims to predict the power consumption in the company's manufacturing plants. The model utilizes sensor data gathered from all of the plants, and these sensors generate tens of millions of records daily. Your objective is to set up a daily training schedule for your model, utilizing all the data collected up to the current date. Additionally, you want the model to scale efficiently with minimal development effort. What steps should you take to achieve this?

  • AA. Train a regression model using Vertex AI AutoML Tables.
  • BB. Develop a custom TensorFlow regression model, and optimize it using Vertex AI Training.
  • CC. Develop a custom scikit-learn regression model, and optimize it using Vertex AI Training.
  • DD. Develop a regression model using BigQuery ML.
  Show answer
• #138

  You need to develop an online model prediction service that accesses pre-computed near-real-time features and returns a customer churn probability value. The features are saved in BigQuery and updated hourly using a scheduled query. You want this service to be low latency and scalable and require minimal maintenance. What should you do?

  • AA. 1. Configure a Cloud Function that exports features from BigQuery to Memorystore. 2. Use Memorystore to perform feature lookup. Deploy the model as a custom prediction endpoint in Vertex AI, and enable automatic scaling.
  • BB. 1. Configure a Cloud Function that exports features from BigQuery to Memorystore. 2. Use a custom container on Google Kubernetes Engine to deploy a service that performs feature lookup from Memorystore and performs inference with an in-memory model.
  • CC. 1. Configure a Cloud Function that exports features from BigQuery to Vertex AI Feature Store. 2. Use the online service API from Vertex AI Feature Store to perform feature lookup. Deploy the model as a custom prediction endpoint in Vertex AI, and enable automatic scaling.
  • DD. 1. Configure a Cloud Function that exports features from BigQuery to Vertex AI Feature Store. 2. Use a custom container on Google Kubernetes Engine to deploy a service that performs feature lookup from Vertex AI Feature Store’s online serving API and performs inference with an in-memory model.
  Show answer
• #139Choose 2

  You're developing a model to enhance your company's online advertising campaigns, aiming to create a dataset for model training while avoiding the creation or reinforcement of unfair bias. What steps should you take? (Choose two.)

  • AA. Include a comprehensive set of demographic features.
  • BB. Include only the demographic groups that most frequently interact with advertisements.
  • CC. Collect a random sample of production traffic to build the training dataset.
  • DD. Collect a stratified sample of production traffic to build the training dataset.
  • EE. Conduct fairness tests across sensitive categories and demographics on the trained model.
  Show answer
• #140

  You work for a company that manages a ticketing platform for a large chain of cinemas. Customers use a mobile app to search for movies they’re interested in and purchase tickets in the app. Ticket purchase requests are sent to Pub/Sub and are processed with a Dataflow streaming pipeline configured to conduct the following steps: 1. Check for availability of the movie tickets at the selected cinema. 2. Assign the ticket price and accept payment. 3. Reserve the tickets at the selected cinema. 4. Send successful purchases to your database. Each step in this process has low latency requirements (less than 50 milliseconds). You have developed a logistic regression model with BigQuery ML that predicts whether offering a promo code for free popcorn increases the chance of a ticket purchase, and this prediction should be added to the ticket purchase process. You want to identify the simplest way to deploy this model to production while adding minimal latency. What should you do?

  • AA. Run batch inference with BigQuery ML every five minutes on each new set of tickets issued.
  • BB. Export your model in TensorFlow format, and add a tfx_bsl.public.beam.RunInference step to the Dataflow pipeline.
  • CC. Export your model in TensorFlow format, deploy it on Vertex AI, and query the prediction endpoint from your streaming pipeline.
  • DD. Convert your model with TensorFlow Lite (TFLite), and add it to the mobile app so that the promo code and the incoming request arrive together in Pub/Sub.
  Show answer

Want to track your score, take timed mock exams and get AI explanations? Create a free account