1. komponenta

Lecture type	Total
Lectures	30
Exercises	15

* Load is given in academic hour (1 academic hour = 45 minutes)

COURSE AIMS AND OBJECTIVES:
Enable students to:
- understand the basic theoretical and practical concepts in the context of deep learning and deep neural networks;
- understand the training process and data needs for deep learning;
- have a working knowledge of deep learning framework for solving realistic problems.

COURSE DESCRIPTION AND SYLLABUS:
1. Introduction: DL versus conventional ML.
2. Deep NN basics: backpropagation, activation functions, output and loss functions.
3. Computation Graphs.
4. Regularization methods for DNNs.
5. Optimization algorithms and strategies.
6. Convolutional Neural Networks.
7. Sequence Modeling Architectures.
8. Graph Neural Networks.
9. Autoencoders.
10. Generative Adversarial Networks.

1. Deep learning, Goodfellow, Ian, Yoshua Bengio, and Aaron Courville, MIT press, 2016.
2. Dive into deep learning, Zhang, Aston, Zachary C. Lipton, Mu Li, and Alexander J. Smola, arXiv preprint arXiv:2106.11342, 2021.

Ostali izborni predmeti - Regular study - Computer Science and Mathematics

Ostali izborni predmeti - Regular study - Computer Science and Mathematics

Ostali izborni predmeti - Regular study - Computer Science and Mathematics

Ostali izborni predmeti - Regular study - Computer Science and Mathematics