School of Engineering \ Computer Engineering
Course Credit
ECTS Credit
Course Type
Instructional Language
Programs that can take the course
Perceptron, Multi-Layer Perceptron, Backpropagation algorithm, Deep Neural Nets (NN): Training, inference, transfer learning, optimizations, Convolutional NNs, Fully Convolutional NNs, Recurrent NNs, LSTMs, Autoencoders, Variational Autoencoders, Generative Adversarial Nets, Transformers, Large Language Models, Stable Diffusion, Normalizing Flow
Textbook and / or References
“Hands-on Machine Learning with Scikit-Learn, Keras & TensorFlow (3rd Ed. )”, by Aurélien Géron
This course provides a comprehensive introduction to deep learning, covering fundamental concepts and advanced techniques in neural networks. Students will explore perceptrons, multi-layer perceptrons, and the backpropagation algorithm before diving into deep neural networks (DNNs) for training, inference, and optimization. The curriculum includes convolutional and fully convolutional networks for image processing, recurrent neural networks (RNNs) and LSTMs for sequential data, and generative models such as autoencoders, variational autoencoders, and generative adversarial networks (GANs). Additionally, students will study transformers, large language models (LLMs), and cutting-edge techniques like stable diffusion and normalizing flows, equipping them with the skills to apply modern AI techniques to real-world problems.
1. Multi layer perceptrons, fully connected layers, deep neural networks and their differences from shallow networks
2. Training deep neural networks, Backpropagation algorithm
3. Convolutional layers, object detection and classification, segmentation
4. Recursive neural networks (RNN), Long Short Term Memory networks, Gated Recurrent Units
5. Embedding spaces, Natural Language Processing with RNNs, Transformer architecture and its applications, MAMBA architecture
6. Auto-encoders, variational auto-encoders, Generative Adversarial Network models, Stable diffusion
Week 1: Multi-layer perceptron
Week 2: Backpropagation algorithm
Week 3: Training deep neural networks
Week 4: Optimizers, transfer learning
Week 5: CNNs
Week 6: RNNs, LSTMs, GRUs
Week 7: RNNs, LSTMs, GRUs
Week 8: Embedding spaces
Week 9: Transformer architecture
Week 10: Variational auto-encoders, GANs
Week 11: Stable diffusion
Week 12: MAMBA architecture
Tentative Assesment Methods
• Midterm 1 25%
• Midterm 2 25%
• Final 35%
• Quizes 15%
|
Program Outcome
**
|
| 1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
11 |
|
Course Outcome
|
|---|
| 1 |
|
|
|
|
|
|
|
|
|
|
|
| 2 |
|
|
|
|
|
|
|
|
|
|
|
| 3 |
|
|
|
|
|
|
|
|
|
|
|
| 4 |
|
|
|
|
|
|
|
|
|
|
|
| 5 |
|
|
|
|
|
|
|
|
|
|
|
| 6 |
|
|
|
|
|
|
|
|
|
|
|