par Saerens, Marco 
;Soquet, Alain
Référence IEE Conference Publication, 313, page (211-215)
Publication Publié, 1989
;Soquet, Alain Référence IEE Conference Publication, 313, page (211-215)
Publication Publié, 1989
Article révisé par les pairs
| Résumé : | We have presented a method for the evaluation of the error to be back-propagated. The method allows a neural net to learn to control a plant in an autonomous way, without a specific learning stage. This evaluation is based on qualitative knowledge of the plant. The field of application of the method is specified. We applied the proposed method to three different problems. These three simulations investigate the possibility of on-line learning with dynamic targets. The results show that dynamic control based on the proposed learning scheme is possible for a neural network. |
