Deep Learning Lab
Through practical exercises, students will not only deepen their understanding of neural network based models, with a wide range of exciting applications, but they will be also exposed to various practical considerations, such as hyper-parameter tuning, which are crucial to make deep learning systems to perform well in practice.
This course will introduce students to practical implementations of various deep learning models using Python and the PyTorch library. Recommended lectures are: Machine Learning, and basic courses on Linear Algebra, Analysis, Probability & Statistics. While it is not a hard requirement, basic knowledge of Python will be greatly helpful.
About 50 percent of the sessions consist of guided exercises. Different programming exercises will cover various building blocks and applications of deep learning. Students will be evaluated through practical assignments.
Students will be evaluated through practical assignments.
- Machine Learning
Useful references (the PDF version of these books can be downloaded free from the following links):
- For machine learning in general:
Bishop, C. M. (2006). Pattern Recognition and Machine Learning. Springer. https://www.microsoft.com/en-us/research/people/cmbishop/prml-book/
- More on recurrent neural networks:
Graves, A. (2012). Supervised Sequence Labelling with Recurrent Neural Networks. Springer. https://www.cs.toronto.edu/~graves/preprint.pdf
- For Python:
Downey, A. B. (2015). Think Python. How to Think Like a Computer Scientist. 2nd Edition. Green Tea Press. http://greenteapress.com/thinkpython2/thinkpython2.pdf