CENG 506

Deep Learning

This course covers methods for designing and training deep neural networks. The course content includes the historical evolution of deep neural networks, their fundamental working principles and image classification and object detection and recognition in images using convolutional neural networks.

Learning Outcomes:

Ability to comment about historical evolution of biometric recognition
Ability to list the basic working principles of artificial neural networks
Ability to apply the learned fundamental principles to train artificial neural networks
Ability to classify images using convolutional neural networks
Ability to apply the acquired knowledge in practice using artificial neural network coding frameworks such as torch, TensorFlow.
Topics
History of artificial neural networks and introduction
Image classification with linear methods – I
Image classification with linear methods – II
Backpropagation – I
Backpropagation – II
Training artificial neural networks I: data preprocessing, weight initialization and regularization, batch normalization and loss functions
Training artificial neural networks II: gradient checking, babysitting the training process, update methods, hyper-parameter optimization
Convolutional neural networks – I
Convolutional neural networks – II
Spatial localization and object detection with convolutional neural networks
Visualization and understanding of convolutional neural networks
Recurrent neural networks
Training convolutional networks in practice: Data augmentation, transfer learning
Deep neural network coding frameworks