DEEP LEARNING AND OPTIMIZATION (4 SCU)

Learning Outcomes:

On successful completion of this course, students will be able to: LO1 – explain the fundamental deep learning theory; LO2 – execute a proper deep learning experimentation workflow; LO3 – analyze architecture of deep learning model; LO4 – compose a deep learning code in Python programming.

Topics:

1. Machine Learning Overview;
2. Multi-layer Perceptrons;
3. Deep Neural Networks;
4. Convolutional Neural Networks;
5. Recurrent Neural Networks;
6. Attention and Memory;
7. Autoencoders and Autoregressive Models;
8. Generative Adversarial Networks;
9. Variational Autoencoders;
10. Self-supervised Learning;
11. Introduction to Deep Learning;
12. Understanding and Visualizing CNN;
13. Project Presentation.