Fundamentals of Statistics contains material of various lectures and courses of H. Lohninger on statistics, data analysis and chemometrics......click here for more.


Natural Brains

Beginners in the field often think that artificial neural networks resemble natural brains. However, there is little similarity between natural brains and the neural networks which are used for data analysis. They are called "neural networks," because the idea to build neural networks stems from natural neural systems. However, when processing data, this origin is not relevant.

In fact neural networks have been designed to follow two main research strains:

  • the modeling of natural brains, and
  • the (non-linear) processing of data.
Cognitive science is concerned with modeling natural brains. The aim is to find out more about how brains work. Thus, biologically plausible models are designed.

When using neural networks for data processing, they are merely employed as statistical tools. Their biological roots can be ignored, because biological plausibility is not relevant.