Deep learning

• Pravila kolegija, uvod u neuronske mreže i duboko učenje
• Matematička osnova, rad i koncepti neuronskih mreža i dubokog učenja
• Osnove programskog jezika visoke razine i alata za rad s podacima
• Programski okviri za pripremu, obradu i vizualizaciju numeričkih i grafičkih tipova podataka
• Programski okviri za oblikovanje i rad s neuronskim mrežama
• Duboki potpuno povezani sekvencijski modeli neuronskih mreža
• Konvolucijski modeli neuronskih mreža
• Kolokvij
• Poteškoće prilikom treniranja neuronskih mreža
• Tehnike treniranja neuronskih mreža
• Tehnike modeliranja neuronskih mreža
• Rekurentne neuronske mreže i rad s kontinuiranim podacima
• Generativni modeli neuronskih mreža
• Napredni modeli neuronskih mreža
• Završni ispit

1. ability to design deep models in a high-level programming language
2. ability to organize and optimize data into structures suitable for training deep neural network models
3. ability to sketch a conceptual deep model of a neural network
4. ability to construct deep neural network models suitable for supervised, semi-supervised and uncontrolled learning
5. ability to disassemble the functional parts of deep neural networks and adapt them to another purpose
6. ability to design a high-performance deep neural network model

Required reading

• Goodfellow, I., Bengio, Y., & Courville, A. (2017). Deep Learning, http://www.deeplearningbook.org
  Aggarwal, C. C. (2018). Neural Networks and Deep Learning. In Springer. https://doi.org/10.1016/B978-0-12-801522-3.00018-5
  Czum, J. M. (2020). Dive Into Deep Learning
  Valkov V. (2020), Get SH*T Done with PyTorch: Solve Real-World Machine Learning Problems with Deep Neural Networks in Python
  Vasilev I., Slater D., Spacagna G., Roelants P., Zocca V. (2019), Python Deep Learning Second Edition, Packt Publishing Ltd.