Incorporating structural plasticity into self-organization recurrent networks for sequence learning
IntroductionSpiking neural networks (SNNs), inspired by biological neural networks, have received a surge of interest due to its temporal encoding. Biological neural networks are driven by multiple plasticities, including spike timing-dependent plasticity (STDP), structural plasticity, and homeostat...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-08-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2023.1224752/full |
| _version_ | 1797767568454320128 |
|---|---|
| author | Ye Yuan Yongtong Zhu Jiaqi Wang Ruoshi Li Xin Xu Tao Fang Hong Huo Lihong Wan Qingdu Li Na Liu Shiyan Yang |
| author_facet | Ye Yuan Yongtong Zhu Jiaqi Wang Ruoshi Li Xin Xu Tao Fang Hong Huo Lihong Wan Qingdu Li Na Liu Shiyan Yang |
| author_sort | Ye Yuan |
| collection | DOAJ |
| description | IntroductionSpiking neural networks (SNNs), inspired by biological neural networks, have received a surge of interest due to its temporal encoding. Biological neural networks are driven by multiple plasticities, including spike timing-dependent plasticity (STDP), structural plasticity, and homeostatic plasticity, making network connection patterns and weights to change continuously during the lifecycle. However, it is unclear how these plasticities interact to shape neural networks and affect neural signal processing.MethodHere, we propose a reward-modulated self-organization recurrent network with structural plasticity (RSRN-SP) to investigate this issue. Specifically, RSRN-SP uses spikes to encode information, and incorporate multiple plasticities including reward-modulated spike timing-dependent plasticity (R-STDP), homeostatic plasticity, and structural plasticity. On the one hand, combined with homeostatic plasticity, R-STDP is presented to guide the updating of synaptic weights. On the other hand, structural plasticity is utilized to simulate the growth and pruning of synaptic connections.Results and discussionExtensive experiments for sequential learning tasks are conducted to demonstrate the representational ability of the RSRN-SP, including counting task, motion prediction, and motion generation. Furthermore, the simulations also indicate that the characteristics arose from the RSRN-SP are consistent with biological observations. |
| first_indexed | 2024-03-12T20:40:43Z |
| format | Article |
| id | doaj.art-003c2f5047324bb9a156e854bb709bb0 |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-03-12T20:40:43Z |
| publishDate | 2023-08-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-003c2f5047324bb9a156e854bb709bb02023-08-01T08:26:42ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2023-08-011710.3389/fnins.2023.12247521224752Incorporating structural plasticity into self-organization recurrent networks for sequence learningYe Yuan0Yongtong Zhu1Jiaqi Wang2Ruoshi Li3Xin Xu4Tao Fang5Hong Huo6Lihong Wan7Qingdu Li8Na Liu9Shiyan Yang10School of Health Science and Engineering, Institute of Machine Intelligence, University of Shanghai for Science and Technology, Shanghai, ChinaSchool of Health Science and Engineering, Institute of Machine Intelligence, University of Shanghai for Science and Technology, Shanghai, ChinaSchool of Health Science and Engineering, Institute of Machine Intelligence, University of Shanghai for Science and Technology, Shanghai, ChinaSchool of Health Science and Engineering, Institute of Machine Intelligence, University of Shanghai for Science and Technology, Shanghai, ChinaSchool of Health Science and Engineering, Institute of Machine Intelligence, University of Shanghai for Science and Technology, Shanghai, ChinaAutomation of Department, Shanghai Jiao Tong University, Shanghai, ChinaAutomation of Department, Shanghai Jiao Tong University, Shanghai, ChinaOrigin Dynamics Intelligent Robot Co., Ltd., Zhengzhou, ChinaSchool of Health Science and Engineering, Institute of Machine Intelligence, University of Shanghai for Science and Technology, Shanghai, ChinaSchool of Health Science and Engineering, Institute of Machine Intelligence, University of Shanghai for Science and Technology, Shanghai, ChinaEco-Environmental Protection Institution, Shanghai Academy of Agricultural Sciences, Shanghai, ChinaIntroductionSpiking neural networks (SNNs), inspired by biological neural networks, have received a surge of interest due to its temporal encoding. Biological neural networks are driven by multiple plasticities, including spike timing-dependent plasticity (STDP), structural plasticity, and homeostatic plasticity, making network connection patterns and weights to change continuously during the lifecycle. However, it is unclear how these plasticities interact to shape neural networks and affect neural signal processing.MethodHere, we propose a reward-modulated self-organization recurrent network with structural plasticity (RSRN-SP) to investigate this issue. Specifically, RSRN-SP uses spikes to encode information, and incorporate multiple plasticities including reward-modulated spike timing-dependent plasticity (R-STDP), homeostatic plasticity, and structural plasticity. On the one hand, combined with homeostatic plasticity, R-STDP is presented to guide the updating of synaptic weights. On the other hand, structural plasticity is utilized to simulate the growth and pruning of synaptic connections.Results and discussionExtensive experiments for sequential learning tasks are conducted to demonstrate the representational ability of the RSRN-SP, including counting task, motion prediction, and motion generation. Furthermore, the simulations also indicate that the characteristics arose from the RSRN-SP are consistent with biological observations.https://www.frontiersin.org/articles/10.3389/fnins.2023.1224752/fullspiking neural networkself-organizationreward-modulated spike timing-dependent plasticityhomeostatic plasticitystructural plasticity |
| spellingShingle | Ye Yuan Yongtong Zhu Jiaqi Wang Ruoshi Li Xin Xu Tao Fang Hong Huo Lihong Wan Qingdu Li Na Liu Shiyan Yang Incorporating structural plasticity into self-organization recurrent networks for sequence learning Frontiers in Neuroscience spiking neural network self-organization reward-modulated spike timing-dependent plasticity homeostatic plasticity structural plasticity |
| title | Incorporating structural plasticity into self-organization recurrent networks for sequence learning |
| title_full | Incorporating structural plasticity into self-organization recurrent networks for sequence learning |
| title_fullStr | Incorporating structural plasticity into self-organization recurrent networks for sequence learning |
| title_full_unstemmed | Incorporating structural plasticity into self-organization recurrent networks for sequence learning |
| title_short | Incorporating structural plasticity into self-organization recurrent networks for sequence learning |
| title_sort | incorporating structural plasticity into self organization recurrent networks for sequence learning |
| topic | spiking neural network self-organization reward-modulated spike timing-dependent plasticity homeostatic plasticity structural plasticity |
| url | https://www.frontiersin.org/articles/10.3389/fnins.2023.1224752/full |
| work_keys_str_mv | AT yeyuan incorporatingstructuralplasticityintoselforganizationrecurrentnetworksforsequencelearning AT yongtongzhu incorporatingstructuralplasticityintoselforganizationrecurrentnetworksforsequencelearning AT jiaqiwang incorporatingstructuralplasticityintoselforganizationrecurrentnetworksforsequencelearning AT ruoshili incorporatingstructuralplasticityintoselforganizationrecurrentnetworksforsequencelearning AT xinxu incorporatingstructuralplasticityintoselforganizationrecurrentnetworksforsequencelearning AT taofang incorporatingstructuralplasticityintoselforganizationrecurrentnetworksforsequencelearning AT honghuo incorporatingstructuralplasticityintoselforganizationrecurrentnetworksforsequencelearning AT lihongwan incorporatingstructuralplasticityintoselforganizationrecurrentnetworksforsequencelearning AT qingduli incorporatingstructuralplasticityintoselforganizationrecurrentnetworksforsequencelearning AT naliu incorporatingstructuralplasticityintoselforganizationrecurrentnetworksforsequencelearning AT shiyanyang incorporatingstructuralplasticityintoselforganizationrecurrentnetworksforsequencelearning |