Abstract
Echo state network (ESN) is a computational model based on reservoir computing framework, but regulation of activation function within the reservoir cannot be done, which affects its nonlinear mapping capability. In this paper, in view of design problem of activation function, we put forward a new activation function design method combining the original function and its derivative functions linearly. In reservoirs, different nodes have different activation functions, which improves the Nonlinear mapping capability of the ESN. Then the outputs of reservoirs are determined by the combined activation functions and the weight of the output layer is obtained by the least square estimation (LSE) method. Finally, simulation results in NARMA and Lorenz systems demonstrate the good effect of the new design method.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
