Artificial Neural Networks

Written by


This course describes the use of neural network models in learning and optimization  e.g., pattern recognition, routing, and prediction. This course is divided into two parts. The first part (3 weeks) provides an introduction to neural networks, focusing on the so-called Hopfield model, its statistical mechanics and optimization algorithms. The second part (4 weeks) provides a more detailed introduction to learning, describing models, algorithms, and applications.

Part 1: An introduction to neural networks: neuroscience, statistics and optimization

  • Neuroscience
  • An example of a neural network: the Hopfield model
  • Statistics and Optimisation (Monte-Carlo methods, simulated annealing)

Part 2: Learning: models, algorithms, and applications

  • Supervised learning: simple perceptrons and layered networks
  • Performance of multilayer perceptrons
  • Unsupervised learning
  • Reinforcement learning
  • Recurrent networks and time series analysis


Student portal page:

External course webpage: None that currently is alive




Bernhard Mehlig

A fundamental question in the statistical physics of  complex systems is how spatial and temporal randomness may generate patterns and dynamics.

The dynamics of complex systems can be systematically analysed using diffusion equations and random-matrix theory.  This approach may yield, as experience shows, surprisingly universal, and in several cases analytically exact results. More importantly, the results show that at first sight unrelated phenomena observed in complex systems in a wide range of different disciplines (Biology, Condensed Matter Physics, and Fluid Dynamics) can be understood in terms of simple and thus general mechanisms. | This email address is being protected from spambots. You need JavaScript enabled to view it.

Latest from Bernhard Mehlig