AWS Certified AI Practitioner (AIF-C01) — Question 268

An AI practitioner is building a model to generate images of humans in various professions. The AI practitioner discovered that the input data is biased and that specific attributes affect the image generation and create bias in the model.
Which technique will solve the problem?

Answer options

• A. Data augmentation for imbalanced classes
• B. Model monitoring for class distribution
• C. Retrieval Augmented Generation (RAG)
• D. Watermark detection for images

Correct answer: A

Explanation

Data augmentation for imbalanced classes helps mitigate bias by artificially balancing the training dataset, ensuring underrepresented attributes or demographics are sufficiently represented. In contrast, model monitoring only tracks distribution issues post-deployment without fixing them, while Retrieval Augmented Generation (RAG) and watermark detection are unrelated to resolving training data demographic bias.