Broad echo state network based on neuronal intrinsic plasticity
TAO Xianfeng
LUN Shuxian
Abstract：The neuronal intrinsic plasticity has been widely used to regulate neuron activity in neural networks,particularly gaining significant advantages in echo state networks.However,existing neuronal intrinsic plasticity typically employ a fixed set of rule parameters for neurons,greatly limiting the network's learning performance.This paper integrates the neuronal intrinsic plasticity with a multi-reservoir echo state network,employing different rule parameters for each reservoir,optimized through a differential evolution algorithm.Experimental results demonstrate that compared to echo state networks without neuronal intrinsic plasticity and those with leaky integrate neurons,the proposed model exhibits superior predictive performance.
Keywords：neuronal intrinsic plasticitybroad echo state networktime series prediction
Publication Date：2024-12-15
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：9( 312-320 )