Deep Learning track

Here we provide a list of topics covered by the Deep Learning track, split into methods and computational aspects. The ordering of topics does not reflect the order in which they will be introduced.

As explained in the overview of courses, the track consists of four levels. If you would like to understand how the curriculum is reflected by each of the four levels, please contact us, and we'll be happy to explain.



Methods


Datasets


Regression problems


Classification problems


Performance evaluation, result significance, and common mistakes

  • Performance metrics and their relationship to the problem’s practical objective
  • Dealing with imbalanced datasets
  • Choosing the right baselines
  • Ablation studies
  • Significance of results

  • Supervised machine learning: basics


    Optimization methods


    Deep Learning libraries

  • Computation graphs
  • Graph-based computations vs. eager execution
  • TensorFlow and Keras
  • PyTorch
  • Production considerations
  • Research considerations

  • Practical aspects of training neural networks


    Practical aspects of training machine learning models


    Improving the performance of machine-learning models


    Neural networks based on fully connected layers


    Convolutional neural networks (CNN)


    Recurrent neural networks (RNN)


    Neural networks with entity embedding layers


    Natural Language Processing (NLP)


    Unsupervised machine learning: basics


    Autoencoders


    Generative Adversarial Networks (GAN)


    Vulnerabilities of machine-learning systems


    Time-series analysis


    Causal inference


    Reinforcement learning


    Benchmarks



    Computational aspects


    Command line interfaces and operating systems


    GPU computing


    Optional: Building one's own GPU-powered computer


    Python: language structure


    Python: libraries


    Python: computation speed


    Data formats


    Docker


    Cloud computing