文献类型：期刊文献

英文题名：Computational Modeling of Structural Synaptic Plasticity in Echo State Networks

作者：Wang, Xinjie[1,2];Jin, Yaochu[2,3];Hao, Kuangrong[2]

机构：[1]East China Univ Sci & Technol, Key Lab Adv Control & Optimizat Chem Proc, Minist Educ, Shanghai 200237, Peoples R China;[2]Donghua Univ, Coll Informat Sci & Technol, Engn Res Ctr Digitized Text & Apparel Technol, Minist Educ, Shanghai 201620, Peoples R China;[3]Univ Surrey, Dept Comp Sci, Guildford GU2 7XH, Surrey, England

年份：2022

卷号：52

期号：10

起止页码：11254

外文期刊名：IEEE TRANSACTIONS ON CYBERNETICS

收录：;EI(收录号：20211310151483);WOS:【SCI-EXPANDED(收录号:WOS:000732297100001)】；

基金：This work was supported in part by the National Natural Science Foundation of China (Basic Science Center Program) under Grant 61988101; in part by the International (Regional) Cooperation and Exchange Project under Grant 61720106008; and in part by the National Natural Science Foundation of China under Grant 61725301 and Grant 61890930-3.

语种：英文

外文关键词：Reservoirs; Neurons; Computational modeling; Synapses; Adaptation models; Task analysis; Training; Classification; echo state networks (ESNs); regression; structural plasticity; structural synaptic plasticity; synaptic plasticity

摘要：Most existing studies on computational modeling of neural plasticity have focused on synaptic plasticity. However, regulation of the internal weights in the reservoir based on synaptic plasticity often results in unstable learning dynamics. In this article, a structural synaptic plasticity learning rule is proposed to train the weights and add or remove neurons within the reservoir, which is shown to be able to alleviate the instability of the synaptic plasticity, and to contribute to increase the memory capacity of the network as well. Our experimental results also reveal that a few stronger connections may last for a longer period of time in a constantly changing network structure, and are relatively resistant to decay or disruptions in the learning process. These results are consistent with the evidence observed in biological systems. Finally, we show that an echo state network (ESN) using the proposed structural plasticity rule outperforms an ESN using synaptic plasticity and three state-of-the-art ESNs on four benchmark tasks.