Skip to content
ml-recall

#nin

3 questions from Landmark Architectures

Timed test

3 questions · 3 min · score and explanations at the end

Take a test
Q 1 / 3Score 0 / 0

Question 1: Short answer

easyAI-checkedLandmark Architectures

NiN's last block outputs a 10×5×510 \times 5 \times 5 tensor, one channel per class, and global average pooling turns it into 10 logits. If you instead flatten the tensor and use a fully connected layer to produce the 10 logits, how many parameters (weights and biases) does that layer add?

a whole number