About the Course

Deep Learning uses a cascade of multiple layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. The Deep Learning (Level 1) has the following contents.

Main Contents

Introduction to Deep Learning
Image Features
Artificial Neural Networks (ANN)
  • ANN architecture and components
  • ANN training process
  • Multi-layer Perceptrons
  • Optimization
  • Stochastic Gradient Descent
  • Backpropagation
  • ANN examples - Voice recognition
Convolutional Neural Networks (CNN)
  • CNN architecture
  • Convolution layer
  • ReLU layer
  • Pooling layer
  • Output layer
  • Designing CNN
  • Training CNN, update rules, data augmentation, ensembles
Popular Convolutional Neural Networks
  • VGG
  • GoogLeNet
  • ResNet
  • DenseNet
Introduction to Open Sources: Keras
Case studies