Deep Learning (OPEN UNI)

The introductory course on deep learning presents the basic concepts, theory, algorithms and models, and provides hands-on experience on implementing, training and utilizing deep neural networks. The topics covered include: linear and logistic regression, loss functions, fully-connected feed-forward neural networks, backpropagation, gradient descent, convolutional neural networks (CNNs), recurrent neural networks (RNNs), attention (including transformers and vision transformers), generative adversarial networks (GANs), variational autoencoder, diffusion models and practical tips for training and utilizing deep neural networks. Various applications of deep learning in fundamental computer vision tasks, such as image classification, object detection and segmentation, are also presented. Finally, limitations, recent progress and new frontiers of deep learning are also discussed.