Informazioni sul corso

This lecture covers advanced topics in deep learning, beginning with its distinctive features within the field machine learning. It explores advanced supervised models like residual networks, transformers, and graph neural networks. We will then go into advanced unsupervised and generative models, including GANs, VAEs, and diffusion models, before concluding with deep reinforcement learning and ethical considerations in deep learning.

After actively participating in this course, learners should be able to

• Define and explain key concepts and models in Deep Learning,
• Explain how to choose a model based on the properties of data in a given problem domain (inductive biases),
• Explain advanced models for major application domains (such as graph learning, generative modelling, unsupervised representation learning, or Reinforcement Learning),
• Critically discuss a scientific publication in the field of Deep Learning,
• Give a scientific presentation,
• Prepare and write a scientific publication. 

The lecture will be held in english. 

See also: 
https://www.mathcs.uni-leipzig.de/studium/studienorganisation/modulangebot