Prerequisites

Data Science & Statistics

The course aims at providing a basic theoretical and practical understanding of neural network approaches. We start with covering the necessary background on traditional artificial neural networks, backpropagation, online learning, and regularization. Then we explain special methods used in deep learning, like drop-out and rectified linear units. We also talk about further advanced topics like convolutional layers, recurrent neural networks, auto-encoders, and generative adversarial networks (GANs).

Our course requires a basic (!) understanding of:

• Linear algebra: vectors, matrices
• Multivariate calculus: derivatives, gradients, integrals
• Probability theory: random variables, distributions, expectation and variance
• Machine Learning! Preferably the lecture Fortgeschrittene Computerintensive Methoden (Computational Methods II) or Predictive Modeling.
• optimization, e.g., Computerintensive Methoden (Computational Methods I) in the statistics master.
• Solid programming knowledge in R or Python.

In addition, our exercises designed to help for practical applications and open-source deep learning libraries.

Programming Language

All of the covered concepts and algorithms are presented independently of any programming language. But of course we also offer a large variety of applied exercises and notebooks. These are currently in R (TensorFlow) and Python (PyTorch).