Question: Explain why examples from an online product catalog can still help a neural network that will later classify photos sent by shoppers on a mobile app.

Sample answer: The two photo sources are not identical, but they can still share useful visual patterns such as object shape, color, and common viewpoints. Because of that overlap, a model can learn from the larger catalog collection and reuse part of what it learned when it later works on shopper photos. Extra catalog images therefore improve training by giving the network more examples of the same underlying objects.

Key points:

• The catalog collection provides more labeled examples.
• Catalog photos and shopper photos share some visual structure.
• Shared structure lets the model transfer what it learned from one source to the other.
• The added training data can improve performance on shopper photos.

Rubric: Full credit for explaining that the two image sources overlap enough for knowledge transfer, that the larger catalog set adds more examples, and that this helps the model perform better on shopper photos without making unsupported claims.