Recap Deep Learning CNN

An image histogram represents:

• The distribution of pixel intensities

• The overall resolution of the image

• The number of colors used in the image

• The brightness of the image

For an image recognition problem (recognizing the type of defect on a microcontroller), which architecture of neural network would be better suited to solve the problem?

• Multi Layer Perceptron

• Convolutional Neural Network

• Recurrent Neural network

• Perceptron

Which of the following is a data augmentation technique used in image recognition tasks?

• Horizontal flipping

• Random cropping

• Random scaling

• Color jittering

In Transfer Learning, when fine-tuning a pre-trained model, which layer is typically trained while keeping the rest frozen?

• The input layer

• The convolutional layers

• The fully connected layers

• The activation layers

Which of the following CNN architectures is considered one of the state-of-the-art models for image classification tasks due to its deep layers and efficiency?

• LeNet-5

• AlexNet

• VGGNet

• ResNet

What is the key advantage of single-stage object detection models compared to two-stage

models?

• Two-stage models require less training data

• Single-stage models are faster

• Single-stage models are more accurate

• Two-stage models are more robust to occlusion

Which of the following options can be used to reduce overfitting in deep learning models?

• Add more data

• Use architecture that generalizes well

• Add regularization

• Increase architectural complexity

Why are activation functions used in neural networks?

• To introduce non-linearity into the model

• To increase the size of the input data

• To normalize the output to zero

• To make the model linear and simple