Turning a VGGNet paper into a key-facts quiz

Select an option to check the answer and explanation immediately.

Q1. Multiple choice

Supplied generated result

Which statement about the VGG network is correct?

1. It used very small 3×3 convolution filters while increasing depth to 16–19 weight layers.2. It used 5×5 convolution filters while limiting depth to 8–10 weight layers.3. It used only 1×1 convolution filters and increased depth to 30 layers.4. It avoided large filters but limited depth to 8–10 weight layers.

Select an option to check the answer and explanation immediately.

Q2. Multiple choice

Supplied generated result

What did the authors conclude about local response normalization in ILSVRC?

1. It improved performance.2. It did not improve performance and increased memory use and computation time.3. It greatly improved both performance and speed.4. It reduced memory use but increased computation time.

Select an option to check the answer and explanation immediately.

Q3. Multiple choice

Supplied generated result

What is the definition of top-5 error in the ILSVRC-2012 classification evaluation?

1. The proportion of images whose ground-truth category is outside the top five predicted categories.2. The proportion of images whose ground-truth category differs from the top-1 prediction.3. The proportion of images whose ground truth appears in the top five predictions.4. The number of images for which every prediction differs from the ground truth.

Select an option to check the answer and explanation immediately.

Q4. Multiple choice

Supplied generated result

What trend in classification error was reported in VGG single-scale evaluation?

1. Error increased from configuration A to configuration E.2. Error generally decreased from 11 weight layers in A to 19 in E and saturated at 19 layers.3. Error barely changed as depth increased.4. Error increased sharply only for configuration E.

Select an option to check the answer and explanation immediately.

Q5. Multiple choice

Supplied generated result

How were pretrained VGG networks used as image features on other datasets?

1. They removed the final fully connected layer and used the 4096-dimensional penultimate activations as image features.2. They used the final convolutional output directly as class probabilities.3. They used the first fully connected layer without removing anything.4. They isolated only the output-layer weights as image features.

Select an option to check the answer and explanation immediately.

Turning a VGGNet paper into a key-facts quiz

Source and generation conditions

How a PDF becomes a quiz

1. Extract PDF content

2. Organize key facts

3. Draft questions

4. Compare with the source

Supplied QuizGen output

Which statement about the VGG network is correct?

What did the authors conclude about local response normalization in ILSVRC?

What is the definition of top-5 error in the ILSVRC-2012 classification evaluation?

What trend in classification error was reported in VGG single-scale evaluation?

How were pretrained VGG networks used as image features on other datasets?

Factual review notes before publication

Create a quiz from a PDF