DEEP LEARNING (2/2 Credits)

Learning Outcomes:

On successful completion of this course, students will be able to: LO1 – execute structured deep learning experimentation workflows, including data preparation, model training, and evaluation; LO2 – construct deep learning models using modern programming tools and frameworks; LO3 – Analyze the behavior and performance of theoretical and practical deep learning models under various training scenarios; LO4 – evaluate the suitability and limitations of different deep learning approaches for solving real-world problems across domains.

Topics:

1. Recurrent Neural Networks – LAB;
2. Unsupervised Learning – LAB;
3. Transformers – LAB;
4. Word Embeddings – LAB;
5. Generative Models – LAB;
6. Graph Neural Networks – LAB;
7. Neural Network Foundations – LAB;
8. Self-Supervised Learning – LAB;
9. Autoencoders – LAB;
10. Convolutional Neural Networks – LAB;
11. Reinforcement Learning – LAB;
12. Regression and Classification – LAB;
13. Deep Learning Foundations;
14. Model Training and Evaluation;
15. Regression and Classification with MLP;
16. Convolutional Neural Networks;
17. Advanced CNNs and Transfer Learning;
18. Autoencoders and Representation Learning;
19. Generative Models: GANs and Diffusion;
20. Sequence Models: RNNs and LSTMs;
21. Attention and Transformers;
22. Self-Supervised and Contrastive Learning;
23. Graph Neural Networks;
24. Reinforcement Learning and Deep Q-Networks;
25. Final Project Presentation.