Introduction to Deep Learning

Welcome to the web page of the class Introduction to Deep Learning at the Technical University of Košice, a course that is taught in the summer semester in the first year of MSc. studies for students of Intelligent Systems. The course is a continuation of the course Neural networks. This web page provides all necessary information and materials for the course.

Join us in our facebook group.

Grading

To successfully pass this course, you have to meet the following requirements:

1. attendance at lectures and labs (3 absences at most)
2. hand in the semestral project (see lower)
3. get at least 21 points during the semester
4. pass the exam (get at least 31 points)

Course plan

	Lectures	Labs	Team project
Week 1 17. 2. - 23. 2.	Indtroduction to Neural Nets with Tensorflow and Keras codes	web deployment tutorial (Andrij David, MSc.)	creating teams, choosing assignments
Week 2 24. 2. - 1. 3.	TBA	basics of Python, Tensorflow and Keras (Ján Magyar, MSc.)
Week 3 2. 3. - 8. 3.	TBA	convolutional NN for detection (Fouzia Adjailia, MSc.)	research report
Week 4 9. 3. - 15. 3.	TBA	Evaluation of NNs (Miroslav Jaščur, MSc.)
Week 5 16. 3. - 22. 3.	TBA	CNN for segmentation and its evaluation (Patrik Sabol, MSc.)	system design and architecture
Week 6 23. 3. - 29. 3.	TBA	RNN for time series and tabural data (Andrij David, MSc.)
Week 7 30. 3. - 5. 4.	TBA	RNN for text processing (Andrij David, MSc.)
Week 8 6. 4. - 12. 4.	TBA	presentation of first versions	proof of concept
Week 9 13. 4. - 19. 4.	TBA	Easter
Week 10 20. 4. - 26. 4.	TBA	generative adversarial networks (Ján Magyar, MSc.)	progress report
Week 11 27. 4. - 3. 5.	TBA	deep reinforcement learning (Lukáš Hruška, MSc.)
Week 12 4. 5. - 10. 5.	TBA	handing in assignments	final version
Week 13 11. 5. - 17. 5.	TBA	—	—

Assignment

During the semester, each student must participate in the completion of an assignment. Assignments are done by teams of three or four students. Each team project must contain the following:

• a specified research goal
• an overview of state-of-the-art solutions
• front-end application deployed on a server
• a trained DL model
• documentation
• a research paper presenting the results of the team project (the paper can be published in academic journals).

Topics:

1. Land cover classification (consultant Patrik Sabol, MSc.)
2. assignment in cooperation with US Steel (consultant Norbert Ferenčík, MSc.)
3. processing the Guatemala dataset (consultant Miroslav Jaščur, MSc.)
4. Neural Code Completion (consultant Andrij David, MSc.)
5. Super-resolution with GANs (consultant Ján Magyar, MSc.)
6. deep reinforcement learning (consultant Lukáš Hruška, MSc.)
7. topic to be announced (consultant Fouzia Adjailia, MSc.)
8. explainable AI (consultant Ivan Čík, MSc.)

** Sign for assignments **

Sources

• Deep Learning Simplified
• Maths of Intelligence
• Convolutional Neural Networks
• Intro to Deep Learning
• Hluboké neuronové sítě