Course Description

Deep Learning course with application on Computer Vision, using the PyTorch Library.

Academic Year

2023-2024

The course will start on January 8th 2024.

Course objectives and outcomes

Objectives: to provide students with the fundamental elements of deep learning, demonstrate its application to computer vision, and form practical skills in implementing and using deep-learning-based systems.

Outcome: An understanding of fundamental concepts and methods of deep learning and its applications, focusing on computer vision, paired with a set of tools and skills required for realizing functioning deep-learning systems.

Required skills

The course will also give the basis to understand the fundamental concepts of machine learning, so knowing them is recommended but optional.

Course Outline

• Introduction to Machine Learning and Deep Learning (1 hour)
• Machine Learning Foundations (3 hours)
• Data Representation with Tensors (3 hours)
• Learning from Tensors: Gradient Descent and Backpropagation (4 hours)
• Designing and Improving Deep-learning Models for classification (3 hours)
• Real-Time Object Detection with YOLO (3 hours)
• Running Scientific Experiments with PyTorch (3 hours)

Course Assessment

The students can decide to take the exam in one of the following two formats (either one or the other):

1 CFU - written examination

2 CFU - development of a project in teams (max 3 people in each group)