Advanced Deep Learning, 5 cr

The contents of the course (lectures and lab sessions) can be summarized to the following bullet points:

• Advanced techniques for recurrent and convolutional neural networks
  
• Sequence-to-sequence modelling and attention mechanisms
• Reconstruction, denosing, and manifold learning with autoencoders
  
• Generative modelling with variational autoencoders and generative adversarial neural networks
• Adversarial training
  
• Self-supervised and representation learning
• Reinforcement learning
• Advanced deep learning applications (e.g. domain adaptation, machine translation, natural language processing, machine vision, and machine listening)
  
• Implementations in popular and open-source deep learning frameworks (e.g. PyTorch)