Deep Learning

In this course, we introduce the principles of deep learning along with their applications in real word problems, aiming in understanding the various Neural Network architectures and their relation to statistical machine learning. In what follows, we emphasize on supervised learning through Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN, GRU, LSTM), and also Autoencoders and Generative Adversarial Networks as techniques for unsupervised learning. Finally, we focus on examples and applications on a wide range of subjects such as image analysis, computer vision and natural language processing.

The goal of this course is to introduce students to the most important deep learning techniques and their application to real-world problems. Students through understanding the various neural networks architectures and their relation to statistical machine learning will gain experience in solving complex modern problems

Successfully completing the course, the student will be in position to:

• Explain fundamental concepts and basic principles of neural networks and machine learning.
• Deploy and parameterize deep learning algorithms per application category.
• Recognize the benefits of each deep learning architecture.

Utilize modern technologies for the implementation and deployment of deep learning algorithms in Python  (Jupyter, Pytorch, Tensorflow, Keras)

Deep Learning, Goodfellow I., Bengio Y., Courville A., MIT Press (free online), 2016.

Neural Networks and Deep Learning: A Textbook, Charu C. Aggarwal, Springer, Cham, 2018.

Written examinations 

Assessment 

Presentation

Lectures in the classroom using multimedia tools (ppt presentations, video, live-stream lectures) Consolidation exercises during the lectures Material in e-class (lectures power-point, suggested exercises) Additional recommended readings