Deep Learning

About

This one-week Deep Learning course covers theoretical and practical aspects, state-of-the-art deep learning architectures and application examples.

The lectures will introduce to students the fundamental building blocks of deep learning methods and the weaknesses and strengths of the different architectures. Students will learn how to tailor a model for a particular application. During tutorials students practice the theory using exercises and have the opportunity to ask for additional explanation for those parts of the material perceived as more difficult. Computer lab sessions aim at making the material come alive and train students in how the methods learnt in class can actually be applied to data. The lab sessions are meant to work on the assignments, such that the students automatically keep up with the material.

Topics covered
Introduction to Deep Learning (High-level definitions of fundamental concepts and first examples)
Deep Learning components (gradient descent models, loss functions, avoiding over-fitting, introducing asymmetry)
Feed forward neural networks
Convolutional neural networks
Embeddings (pre-trained embeddings, examples of pre-trained models, e.g., Word2Vec)
Generative Adversarial Network (GAN)
Advanced architectures (Densely connected networks, Adaptive structural learning)

Course leader

Eran Raviv holds a PhD in econometrics from Erasmus University Rotterdam, a master’s degree in applied statistics from Tel Aviv University and a second master’s degree in quantitative finance from Erasmus University Rotterdam.

Target group

Level
The summer course welcomes Master’s and PhD students, alumni, professionals in economics and related fields, who are interested in deep learning. The level is introductory, targeted at participants who would like to familiarize themselves with the topic, and acquire a good basis from which to approach deep learning potential applications.

Admission requirements
Students are expected to have a solid background in calculus, linear algebra, and classical statistics. Familiarity with open source languages such as R or Python is a must.

Course aim

Participants who joined at least 80% of all sessions and hand in the assignment will receive a certificate of participation stating that the summer school is equivalent to a work load of 3 ECTS. Note that it is the student’s own responsibility to get these credits registered at their university.

Fee info