CNN Transfer Learning with CNNs 5

Question 1

Which transfer learning strategy is best when the target dataset is small and similar to the source dataset?

Accepted Answer

Freeze the convolutional base, train only the classifier

Answer

Fine-tune all layers with high learning rate

Answer

Train from scratch on target data only

Answer

Use the pretrained model without any modification

Question 2

What is the role of Global Average Pooling (GAP) in modern transfer learning architectures?

Accepted Answer

Reduces each feature map to a single value, replacing flatten layers

Answer

Increases spatial resolution before the classifier

Answer

Performs data augmentation spatially

Answer

Computes attention weights across feature maps

Question 3

In few-shot learning via transfer learning, what enables a CNN to generalize from very few examples?

Accepted Answer

Rich feature representations learned from large source datasets

Answer

Random weight initialization for each class

Answer

Using larger batch sizes during meta-training

Answer

Increasing the number of convolutional filters

Question 4

When applying transfer learning to object detection (e.g., using Faster R-CNN), what part of the network is typically pretrained?

Accepted Answer

The backbone CNN used as a feature extractor

Answer

The region proposal network (RPN)

Answer

The classification head for bounding boxes

Answer

The non-maximum suppression module

Question 5

Which technique involves training a lightweight student model to mimic a larger pretrained teacher model?

Accepted Answer

Knowledge distillation

Answer

Progressive fine-tuning

Answer

Multi-task learning

Answer

Model pruning

Question 6

What is multi-task learning in the context of CNN transfer learning?

Accepted Answer

Training a single CNN on multiple related tasks simultaneously to share representations

Answer

Running multiple pretrained models in parallel

Answer

Sequentially fine-tuning on different datasets

Answer

Using multiple optimizers for different layers

Question 7

Why is batch normalization sometimes problematic when fine-tuning a pretrained CNN on a small dataset?

Accepted Answer

Small batch statistics diverge from pretrained population statistics, causing instability

Answer

It increases the number of trainable parameters significantly

Answer

It prevents gradients from flowing to earlier layers

Answer

Batch normalization layers cannot be frozen