You are here: Home :: Area NEM :: Computer science

Joachim E. Setzer, Eberhard Kienzle

Interpolating Counterpropagation Network for Adaptive Modelling of Characteristics

In theory, neural networks provide a good method for adaptive modelling of non-linear characteristics. However, they haven´t been used in practical applications since most standard networks cannot fulfil the requirements sufficiently. There is a need for a neural network that interpolates continiously based on samples of the characterisic. Furthermore it has to provide the capability of quick re-learning and therefore local adaption. A survey of potential solutions suggests an "Interpolating Counterpropagation Network". The following is a presentation of a realization which fulfils all requirements mentioned above.

it – Information Technology (vormals it+ti), Oldenbourg Wissenschaftsverlag

Print ISSN: 1611-2776
Volume: 40, 01/1998
Pages: 20

Journal homepage (external site)

Show all available items of this journal

 

Area NEM