Task 1.2: Select and Configure FMs. AIP-C01 Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   A developer wants to allow switching between different foundation models without changing application code. Which AWS service is designed to store and manage configuration data for this purpose?

   • AAmazon API Gateway
   • BAWS AppConfig
   • CAmazon S3
   • DAWS Lambda
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   Which technique is a parameter-efficient fine-tuning method that trains only a small set of new parameters while keeping the original model weights frozen?

   • AFull fine-tuning
   • BLoRA (Low-Rank Adaptation)
   • CPrompt engineering
   • DModel distillation
   Reveal Solution Discussion Select an answer first
4. 8expert · medium

   A media streaming company uses Amazon Bedrock to generate personalized recommendations. They have deployed their application in the us-west-2 Region and want to ensure high availability. They are considering using Amazon Bedrock Cross-Region Inference. However, they also have a compliance requirement that data must not leave the United States. Which additional consideration should they address?

   • AEnsure that the secondary Region used by Cross-Region Inference is also within the United States.
   • BImplement a circuit breaker in AWS Step Functions to fail over to a different model in the same Region.
   • CUse Amazon SageMaker to deploy a custom model in a US Region and route traffic to it using Route 53.
   • DUse AWS AppConfig to restrict the model selection to US-based models only.
   Reveal Solution Discussion Select an answer first
5. 9application · medium

   A healthcare company wants to fine-tune a foundation model to improve the accuracy of medical question answering. They have a limited amount of labeled medical data and need to deploy the model for inference. They are considering using LoRA. What is the primary benefit of using LoRA in this scenario?

   • AIt reduces the amount of labeled data required for fine-tuning.
   • BIt improves the model's performance on the fine-tuning task compared to full fine-tuning.
   • CIt reduces the computational resources required for fine-tuning.
   • DIt allows the model to be deployed on a serverless platform without a dedicated endpoint.
   Reveal Solution Discussion Select an answer first
6. 10expert · medium

   A financial services company uses Amazon Bedrock to process loan applications. They have deployed their application in the us-east-1 Region and want to ensure high availability. They are considering using Amazon Bedrock Cross-Region Inference. However, they also have a requirement to minimize latency for their users, who are primarily located on the East Coast of the US. Which additional measure should they implement?

   • AUse Amazon SageMaker to deploy a custom model in the same Region and use a circuit breaker to fail over to it.
   • BUse Amazon Route 53 with a latency-based routing policy to route users to the nearest Region.
   • CUse Amazon Bedrock Cross-Region Inference and configure the secondary Region to be us-east-2, which is also on the East Coast.
   • DUse AWS AppConfig to configure the secondary Region to be us-west-2, which is geographically distant but provides redundancy.
   Reveal Solution Discussion Select an answer first