Adaptive Implementation of Artificial Neural Networks Reflecting Changing Hardware Resources at Run-Time

U. Seiffert (Germany)


Artificial Neural Networks; parallel hardware; changing hardware resources; self-adaptation; Organic Computing.


A number of basic mathematical algorithms along with their ability of a hardware adaptive implementation actu ally already paved the way towards more general and also more complex frameworks. However, artificial neural net works, which often become rather complex and which are inherently suitable and at least when applied to large scale data sets often required to be run on parallel hard ware, have not moved into the focus of real hardware adap tive implementations yet. This paper systematically re views the state-of-the-art and the requirements of imple menting artificial neural networks on varying parallel com puter hardware. Based on this it provides perspectives which clearly extend recent attempts of hardware adaptive implementations based on generally varying but at run-time fixed resources.

Important Links:

Go Back