Menu
A+ A A-

SDS 418: Deep Learning Approaches

Course Title

Deep Learning Approaches

Course Code

SDS 418

Course Type

Elective

Level

Master’s

Year / Semester

1st / 2nd (subject to change)

Instructor’s Name

Mihalis Nicolaou 

ECTS

10

Lectures / week

1 (2h)

Laboratories / week

1 (1h)

Course Purpose and Objectives

To teach students the theoretical concepts on deep learning and how to implement and use them to automatically extract features from data and build prediction models for several applications. Methods covered include feedforward, convolutional, recurrent and recursive networks. 

Learning Outcomes

Students will learn the fundamentals, implement and use deep learning methods for a wide range of applications such as object recognition in images, anomaly detection, pattern recognition in omics (e.g., genomics) data, medical diagnosis, etc. 

Prerequisites

None

 Requirements  -
Course Content

Neural networks: feedforward networks, radial basis function networks, recurrent neural networks, modular neural networks.

Deep learning neural networks: Feedforward networks (autoencoders, restricted Boltzmann machines), convolutional networks, recurrent networks (long short-term memory), recursive networks (recursive autoencoders, recursive neural tensor networks).

Usage and hands on experience of state of the art libraries for Deep Learning.

Teaching Methodology

Lectures, exercises, seminars, reports

Bibliography

1) I. Goodfellow, Y. Bengio A. Courville, “Deep Learning”, MIT press.

2) Sebastian Raschka, David Julian, John Hearty, “Python: Deeper Insights into Machine Learning”, Packt Publishing.

Assessment

25% coursework, 75% exam

Language

English