Content, progress and pedagogy of the module

Learning objectives

Knowledge

• Have comprehension of the benefits of using deep learning models in its applications
• Have comprehension of the supervised, semi-supervised, unsupervised and reinforcement deep learning-based models
• Have fundamental knowledge on optimization techniques and the cost functions used in deep learning systems
• Have fundamental knowledge on transformer and generative based deep learning models
• Have fundamental knowledge on deep neural networks and computer vision

Skills

• Be able to configure deep neural networks for training and inferencing
• Be able to train a deep neural network on highly parallel architectures
• Be able to select the appropriate hardware to perform efficient inferences
• Be able to use deep learning models in simulation environments  
• Be able to analyse the architecture of deep neural networks and their multiple layers
• Be able to analyse typical case studies in the application of deep learning methods in domains such as computer vision and robotics
• Be able to select the appropriate deep learning model for a given application
• Be able to evaluate the performance of deep learning models

Competences

• Be able to design and implement deep learning models in computer vision and robotics for industrial applications
• Be able to use and apply deep learning frameworks to design deep learning models
• Be able to evaluate different deep learning models within an application domain
• Be able to adapt deep learning models to different domains

Type of instruction

The course will be taught by a mixture of lectures, workshops, exercises, mini-projects, self-study and e-learning.

Extent and expected workload

Since it is a 5 ECTS course module, the workload is expected to be 150 hours for the student.

Exam

Exams