Data Science and Machine Learning (Theory and Projects) A to Z - Classical CNNs: LeNet

The video tutorial covers the evolution of convolutional neural networks (CNNs), starting with the pioneering work of John Lee Kuhn and the development of LeNet for digit recognition. It explains the architecture of LeNet, its components, and its impact on machine learning, particularly in ATM machines. The tutorial concludes with a brief introduction to modern CNN architectures like AlexNet, VGG, Inception, and ResNet.

Who was the pioneer behind the first convolutional neural network that gained significant attention?

• Yann LeCun

• Geoffrey Hinton

• Andrew Ng

• John Lee Kuhn

What was the primary purpose of the LeNet architecture when it was first introduced?

• Object detection

• Handwritten digit recognition

• Speech recognition

• Image classification

In the LeNet architecture, what was the size of the filters used in the first convolutional layer?

• 9x9

• 7x7

• 5x5

• 3x3

What nonlinearity was initially used in the LeNet architecture?

• Leaky ReLU

• Sigmoid

• Tanh

• ReLU

How many fully connected layers are present in the LeNet architecture?

• Four

• Three

• One

• Two

What was one of the practical applications of LeNet in the 1990s?

• Weather prediction

• ATM deposit processing

• Speech synthesis

• Facial recognition

Which modern architectures are mentioned as successors to LeNet?

• ResNet and VGG

• AlexNet and Inception

• Both A and B

• None of the above