Deep Learning CNN Convolutional Neural Networks with Python - AlexNet

The video introduces convolutional neural networks (CNNs), focusing on AlexNet, a groundbreaking model that won the ImageNet challenge in 2012. It compares AlexNet with LeNet, highlighting differences in architecture, such as the use of ReLU activation and max pooling in AlexNet. The video emphasizes AlexNet's impact on the field, encouraging a shift from manually designed features to learned features.

What was a significant achievement of AlexNet in 2012?

• Being the first neural network for digit recognition

• Introducing the concept of pooling layers

• Winning the ImageNet challenge by a large margin

• Using sigmoid activation functions

What is the size of the input image for LeNet?

• 28x28

• 32x32

• 224x224

• 64x64

How many fully connected layers does LeNet have?

• Four

• Three

• Two

• One

What is the size of the first convolutional filter in AlexNet?

• 5x5

• 3x3

• 7x7

• 11x11

How many filters are used in the second convolutional layer of AlexNet?

• 384

• 256

• 512

• 96

Which activation function is used throughout AlexNet?

• Softmax

• ReLU

• Tanh

• Sigmoid

What type of pooling is predominantly used in AlexNet?

• Stochastic pooling

• Average pooling

• Global pooling

• Max pooling