1. Which of the following is a best practice for securing data stored in Amazon S3 for machine learning workloads?

A) Use the default S3 bucket policy to grant access to specific IAM users and roles, ensuring that only authorized users can access the data.
B) Use Amazon S3 Access Points to restrict access to specific parts of your S3 buckets based on the user's IAM role, making it easier to control access to data.
C) Use Amazon S3 Transfer Acceleration to improve the speed of data transfers to and from your S3 buckets for machine learning workloads.
D) Enable public access to your S3 buckets to facilitate data sharing for machine learning workloads.
E) Store your data in S3 Glacier storage class to reduce costs and improve data durability for machine learning workloads.


2. Which of the following instance types is best suited for memory-intensive workloads in Amazon EC2?

A) C5
B) M5
C) G4dn
D) T3
E) R5


3. Which of the following neural network architectures is best suited for natural language processing (NLP) tasks that involve learning the sequence of words in a text data?

A) Convolutional Neural Networks (CNN)
B) None of the above
C) Deep Belief Networks (DBN)
D) Autoencoders
E) Recurrent Neural Networks (RNN)


4. You are working on a machine learning project that requires the use of a custom algorithm. You have implemented the algorithm in Python and want to deploy it on AWS. Which of the following services would be the best choice for deploying a custom algorithm for use in a production environment?

A) AWS Elastic Beanstalk
B) AWS Lambda
C) AWS Batch
D) Amazon SageMaker



5. Which of the following statements is true about AWS Glue?

A) AWS Glue is a fully managed extract, transform, and load (ETL) service that makes it easy to move data between data stores.
B) AWS Glue provides a real-time streaming data processing engine that can ingest and process large amounts of data in real-time.
C) AWS Glue is primarily used for building and training machine learning models on large datasets.
D) AWS Glue is an open-source distributed big data processing framework that can process large volumes of data efficiently.