Create a version control system in Git for the model’s training code and configuration files, while storing the trained models in a separate S3 bucket for easy retrieval - Using Git for version control of the training code and configurations is a good practice, but it does not address the need to manage the actual trained models and their associated metadata systematically. The SageMaker Model Registry offers a more comprehensive solution that integrates both code and model versioning. Use SageMaker Model Monitor to track the performance of models in production, ensuring that any changes in model behavior are documented for future audits - SageMaker Model Monitor is useful for monitoring model performance in production, but it does not handle version control or track the metadata necessary for repeatability and audits. It is complementary to, but not a substitute for, the SageMaker Model Registry. References: https://docs.aws.amazon.com/sagemaker/latest/dg/model -registry.html https://docs.aws.amazon.com/sagemaker/latest/dg/model -monitor.html You are a machine learning engineer at a financial services company tasked with building a real -time fraud detection system. The model needs to be highly accurate to minimize false positives and false negatives. However, the company has a limited budget for cloud resources, and the model needs to be retrained frequently to adapt to new fraud patterns. You must carefully balance model performance, training time, and cost to meet these requirements. Which of the following strategies is the MOST LIKELY to achieve an optimal balance between model performance, training time, and cost?
Use a deep neural network with multiple layers and complex architecture to maximize performance, even if it requires significant computational resources and longer training times - A deep neural network may provide high accuracy but typically requires significant computational resources and longer training times, leading to higher costs. This approach may not be feasible within a limited budget, especially with the need for frequent retraining. Deploy a simpler model like logistic regression to reduce training time and cost, while accepting a slight reduction in model accuracy - Logistic regression is simple and cost -effective but may not achieve the level of accuracy required for a critical application like fraud detection. This tradeoff might be too significant if accuracy is compromised. Choose a support vector machine (SVM) with a nonlinear kernel to enhance accuracy, regardless of the increased training time and cost associated with large datasets - SVMs with nonlinear kernels can be very accurate but are computationally intensive, particularly with large datasets. The increased training time and cost might outweigh the benefits, especially when there are more cost -effective alternatives like XGBoost. Reference: https://docs.aws.amazon.com/sagemaker/latest/dg/algos.html You are an ML Engineer working for a healthcare company that uses a machine learning model to recommend personalized treatment plans to patients. The model is deployed on Amazon SageMaker and is critical to the company's operations, as any incorrect predictions could have significant consequences. A new version of the model has been developed, and you need to deploy it in production. However, you want to ensure that the deployment process is robust, allowing you to quickly roll back to the previous version if any issues arise. Additionally, you need to maintain version control for future updates and manage traffic between different model versions. Which of the following strategies should you implement to ensure a smooth and reliable deployment of the new model version using Amazon SageMaker, considering best practices for versioning and rollback strategies? (Select two)
Deploy the new model version immediately and redirect 100% of traffic to it, assuming it has been thoroughly tested and will not require a rollback - Redirecting 100% of traffic to the new model version immediately is risky, especially in a critical application like healthcare. Without a rollback plan, any issues with the new version could lead to significant consequences. Create a backup of the current model, deploy the new version, and if any issues arise, manually roll back by redeploying the previous model version - While creating a backup and manually rolling back is a possible strategy, it is not as efficient or reliable as using a built -in rollback feature like blue/green deployment or canary releases. Manual rollbacks can lead to delays and increased downtime. Deploy the new model version alongside the current one, and use Amazon SageMaker’s multi - model endpoint to serve both models simultaneously, splitting traffic between them - While using a multi -model endpoint could serve both models simultaneously, it is not the best approach for managing risk during deployment. This strategy is more suited for scenarios where you need to serve multiple models for different purposes rather than managing a controlled rollout. References: https://docs.aws.amazon.com/sagemaker/latest/dg/deployment -guardrails -blue-green.html https://docs.aws.amazon.com/sagemaker/latest/dg/deployment -guardrails -rolling.html https://docs.aws.amazon.com/sagemaker/latest/dg/multi -model -endpoints.html You are a data scientist at a marketing agency tasked with creating a sentiment analysis model to analyze customer reviews for a new product. The company wants to quickly deploy a solution with minimal training time and development effort. You decide to leverage a pre -trained natural language processing (NLP) model and fine-tune it using a custom dataset of labeled customer reviews. Your team has access to both Amazon Bedrock and SageMaker JumpStart. Which approach is the MOST APPROPRIATE for fine-tuning the pre-trained model with your custom dataset?
applications. However, as of now, Bedrock does not directly support fine-tuning these models within its interface. Fine -tuning is better suited for SageMaker JumpStart in this scenario. Use Amazon Bedrock to train a model from scratch using your custom dataset, as Bedrock is optimized for training large models efficiently - Amazon Bedrock is not intended for training models from scratch, especially not for scenarios where fine-tuning a pre-trained model would be more efficient. Bedrock is optimized for deploying and scaling foundation models, not for raw model training. Use SageMaker JumpStart to create a custom container for your pre -trained model and manually implement fine-tuning with TensorFlow - While it’s possible to create a custom container and manually fine-tune a model, SageMaker JumpStart already offers an integrated solution for fine -tuning pre -trained models without the need for custom containers or manual implementation. This makes it a more efficient and straightforward option for the task at hand. Reference: https://docs.aws.amazon.com/sagemaker/latest/dg/jumpstart -fine-tune.html You are a machine learning engineer at a healthcare company responsible for developing and deploying an end -to-end ML workflow for predicting patient readmission rates. The workflow involves data preprocessing, model training, hyperparameter tuning, and deployment. Additionally, the solution must support regular retraining of the model as new data becomes available, with minimal manual intervention. You need to select the right solution to orchestrate this workflow efficiently while ensuring scalability, reliability, and ease of management. Given these requirements, which of the following options is the MOST SUITABLE for orchestrating your ML workflow?
Incorrect options: Use AWS Step Functions to define and orchestrate each step of the ML workflow, integrate with SageMaker for model training and deployment, and leverage AWS Lambda for data preprocessing tasks - AWS Step Functions is a powerful service for orchestrating workflows, and it can integrate with SageMaker and Lambda. However, using Step Functions for the entire ML workflow adds complexity since it requires coordinating multiple services, whereas SageMaker Pipelines provides a more seamless, integrated solution for ML -specific workflows. Leverage Amazon EC2 instances to manually execute each step of the ML workflow, use Amazon RDS for storing intermediate results, and deploy the model using Amazon SageMaker endpoints - Manually managing each step of the ML workflow using EC2 instances and RDS is labor - intensive, prone to errors, and not scalable. It also lacks the automation and orchestration capabilities needed for a robust ML workflow. Use AWS Glue for data preprocessing, Amazon SageMaker for model training and tuning, and manually deploy the model to an Amazon EC2 instance for inference - While using AWS Glue for data preprocessing and SageMaker for training is possible, manually deploying the model on EC2 lacks the orchestration and management features provided by SageMaker Pipelines. This approach also misses out on the integrated tracking, automation, and scalability features offered by SageMaker Pipelines. Use AWS Glue for data preprocessing, Amazon SageMaker for model training and tuning, and manually deploy the model to an Amazon EC2 instance for inference - While using AWS Glue for data preprocessing and SageMaker for training is possible, manually deploying the model on EC2 lacks the orchestration and management features provided by SageMaker Pipelines. This approach also misses out on the integrated tracking, automation, and scalability features offered by SageMaker Pipelines. Reference: https://docs.aws.amazon.com/sagemaker/latest/dg/pipelines.html You are a data scientist at an insurance company developing a machine learning model to predict the likelihood of claims being fraudulent. The company has a strong commitment to fairness and wants to ensure that the model does not disproportionately affect any specific demographic group. You decide to use Amazon SageMaker Clarify to assess potential bias in your model. In particular, you are interested in understanding how the model’s predictions differ across demographic groups when conditioned on relevant factors like income level, which could influence the likelihood of fraudulent claims. Given this scenario, which of the following BEST describes how Conditional Demographic Disparity (CDD) can be used to assess and mitigate bias in your model?
➢ TOTAL QUESTIONS:125 You are a machine learning engineer at a fintech company tasked with developing and deploying an end-to-end machine learning workflow for fraud detection. The workflow involves multiple steps, including data extraction, preprocessing, feature engineering, model training, hyperparameter tuning, and deployment. The company requires the solution to be scalable, support complex dependencies between tasks, and provide robust monitoring and versioning capabilities. Additionally, the workflow needs to integrate seamlessly with existing AWS services. Which deployment orchestrator is the MOST SUITABLE for managing and automating your ML workflow?
