Neural networks 2
About the last week of the semester and the student presentations
-
The presentations of the students projects will take place during the lessons in the last week of the semester
(unless a student chooses to present earlier).
Please plan for approximately 15 minutes per presentation.
If someone would like to present for a longer time, this is possible upon prior agreement, preferably with an earlier presentation date.
-
The make-up class for the session cancelled on 28 October 2025 (public holiday) will take place on
16 December 2025 from 12:00 to 13:00 (most probably in room T-115).
Both the regular class and the make-up session will be dedicated to student presentations.
12th week (December 16th 2025)
- Presentations of student projects
11th week (December 9th 2025)
- Natural language processing. Text classification (sentiment analyzis). Introducton to transforers.
- Brief introduction to generative models for imae generation.
- week11.pdf
- GitHub
10th week (December 2nd 2025)
9th week (November 25th 2025)
- Applications of CNNs: object detection,... - finishin the chapter.
- Sequences, Time series and Recurrent neural networks.
- Graded Homework - Email your solution notebook by Dec 8, 2025. Consultation (lab or individual) required by Dec 12, 25 to recieve points.
See slides for details.
- week09.pdf
- GitHub - 1st part
- GitHub - 2nd part
8th week (November 18th 2025)
- Important: 18th November is the deadline for the selection of the students projekts tasks and their approval.
- Convolutional Neural Networks (CNNs): modern architecture patterns - practical examples.
- Applications of CNNs: segmentation, object detection,...
- Graded Homework - Email your solution notebook by Nov 24, 2025. Consultation (lab or individual) required by Nov 28, 25 to recieve points.
See slides for details.
- week08.pdf
- GitHub - 1st part
- GitHub - 2nd part
7th week (November 11th 2025)
- Convolutional Neural Networks (CNNs): transfer learning, modern architecture patterns.
- Graded Homework - Email your solution notebook by Nov 17, 2025. Consultation (lab or individual) required by Nov 21, 25 to recieve points.
See slides for details.
- week07.pdf
- GitHub - 1st part
- GitHub - 2nd part
6th week (November 4th 2025)
- Convolutional Neural Networks (CNNs): introduction, bipyramidal architecture, visualization, training a model from scratch, data loaders, data augmentation, regularization.
- Practical example: CNN on Fashion MMNIST, CNN on a small dataset (Cats and Dogs)
- Graded Homework - Email your solution notebook by Nov 10, 2025. Consultation (lab or individual) required by Nov 14, 25 to recieve points.
See slides for details.
- week06.pdf
- GitHub
5th week (October 21th 2025)
- Neural networks and generalization. techniques for preventing overfitting.
- Introduction to Convolutional Neural Networks
- Graded Homework - Email your solution notebook by Nov 3, 2025. Consultation (lab or individual) required by Nov 7, 25 to recieve points.
See slides for details.
- week05.pdf
- GitHub
4th week (Octber 14th 2025)
- Training workflow in Keras: A complete example.
- Examples of solving various tasks using MLP (classification, regression, time series prediction). Text, image and numerical data.
- Graded Homework - Email your solution notebook by Oct 21, 2025. Consultation (lab or individual) required by Oct 24, 25 to recieve points.
See slides for details.
- week4.pdf
- GitHub
3rd week (Octber 7th 2025)
- Multi-Layer Neural Netwoks and their training.
- Introduction to Python libraries for machine learning and deep learning.
- Training workflow in Keras: A complete example
- Graded Homework - Email your solution notebook by Oct 14, 2025. Consultation (lab or individual) required by Oct 17, 25 to recieve points.
See slides for details.
- week3.pdf
- GitHub
2nd week (September 30th 2025)
- Introduction to Machine Learning and Artificial Neural Networks.
- History of Artificial Neural Networks
- Artificial Neurons
- Multi-Layer Neural Netwoks and their training.
- week2.pdf
- GitHub
1st Week (September 23th 2025)
- About the Course. Overview. Conditions for Credits.
- Introduction to Machine Learning and Artificial Neural Networks.
- Workflow of a Machine Learning Task.
- History of Artificial Neural Networks - next week.
- Introduction to Python libraries for machine learning and deep learning.
- week1.pdf
- GitHub