Fischer, J.J.Fischer2022-03-102022-03-102003https://publica.fraunhofer.de/handle/publica/34388010.1007/3-540-44868-3_39Recurrent neural networks are still a challenge in neural investigation. Most commonly used methods have to deal with several problems like local minima, slow convergence or bad learning results because of bifurcations through which the learning system is driven. The following approach, which is inspired by Echo State networks [1], overcomes those problems and enables learning of complex dynamical signals and tasks.enecurrent neural networksEcho Statelearning005006629400conference paper