Objective 4.2 Secure a Pipeline (includes Maintenance) AIP-210 Practice Questions (Page 1)

Questions 1–5

1. 1foundation · easy

   What is the primary purpose of multi-factor authentication (MFA) for pipeline users?

   • ATo replace the need for role-based access control
   • BTo add an extra layer of security beyond just a password
   • CTo encrypt all data in the pipeline
   • DTo speed up the login process for frequent users
   Reveal Solution Discussion Select an answer first
2. 2foundation · easy

   Which security measure is specifically designed to protect data while it is being transmitted between pipeline components?

   • ARole-based access control on storage
   • BData masking of sensitive fields
   • CEncryption at rest using AES-256
   • DEncryption in transit using TLS
   Reveal Solution Discussion Select an answer first
4. 3foundation · easy

   Which of the following is a key purpose of logging in an ML pipeline?

   • ATo provide an audit trail for security and compliance investigations
   • BTo replace the need for access control
   • CTo automatically retrain models when accuracy drops
   • DTo increase the speed of model training
   Reveal Solution Discussion Select an answer first
5. 4application · medium

   A media company's ML platform uses a shared Azure Machine Learning workspace for multiple project teams. The security team wants to ensure that Team A cannot access Team B's experiment runs or models, while still allowing both teams to use the same compute cluster. What is the most appropriate way to isolate the teams within the shared workspace?

   • ACreate a separate Azure Machine Learning workspace for each team and delete the shared compute cluster.
   • BUse Azure AD groups for each team and assign both groups the Reader role on the workspace, then use a shared key for compute access.
   • CGrant both teams Contributor access to the entire workspace and rely on naming conventions to distinguish their assets.
   • DCreate separate Azure RBAC roles for each team and assign them only to their respective resource groups or sub-resources, while keeping the compute cluster in a shared resource group with its own access role.
   Reveal Solution Discussion Select an answer first
6. 5foundation · easy

   Which of the following is a common security risk associated with the data ingestion stage of an ML pipeline?

   • AModel inversion attacks that reconstruct training data
   • BUnencrypted model files being downloaded by unauthorized users
   • CExcessive permissions granted to the model registry
   • DUntrusted data sources introducing malicious or poisoned data
   Reveal Solution Discussion Select an answer first