Deep Learning
3 hp
Date: 3-5 May 2023 in Lund
Registration deadline: March 31
Lecturers: Amandus Krantz and Christian Balkenius
Course Description
The course Deep Learning for non-programmers introduces Deep Learning on a conceptual and intuitive level, without getting tangled in technical implementation details such as mathematics or programming. The course assumes no mathematical or programming knowledge of the students.
The aim of the course is to provide non-technical students with (1) the knowledge to participate in discussions on Deep Learning and (2) a high-level overview of the field, without necessarily having to learn how to build and develop Deep Learning models.
Topics that will be covered in the course:
– Anatomy of the Artificial Neural Network (ANN)
– Relationship between Deep Learning and Neuroscience
– Training of deep neural networks
– Supervised and unsupervised learning
– Convolutional Neural Networks (CNN)
– Reinforcement learning
– Algorithmic bias
– Applications of deep learning
After the course, the student will:
– Be able to explain and discuss basic deep learning concepts
– Have a conceptual understanding of basic deep learning
Schedule
3 May
13.15-14.00 Introduction (AK/CB)
14.15-15.00 What is an artificial neural network? 1 (AK)
COFFEE
15.15-16.00 What is an artificial neural network? 2 (AK)
16.15-17.00 Fooling a network (CB)
DINNER
4 May
10.15-11.00 Convolutional networks 1 (AK)
11.15-12.00 Convolutional networks 2 (AK)
LUNCH
13.15-14.00 Reinforcement learning 1 (CB)
14.15-15.00 Reinforcement learning 2 (CB)
COFFEE
15.15-16.00 Network architectures 1 (AK)
16.15-17.00 Network architectures 2 (AK)
5 May
9.15-10.00 Bias (CB)
COFFEE
10.15-11.00 Deep learning applications(AK)
11.15-12.00 Conclusion (AK/CB)
19-23 May
Peer reviewing
26 May
Deadline for examination paper
5 pages
Teachers:
Christian Balkenius (CB)
Amandus Krantz (AK)
Registration
Please register no later than March 31! After this date we cannot guarantee a room at Hotel Lundia.