• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

使用遗传算法对模拟多室神经元进行参数化。

Parametrizing analog multi-compartment neurons with genetic algorithms.

作者信息

Stock Raphael, Kaiser Jakob, Müller Eric, Schemmel Johannes, Schmitt Sebastian

机构信息

Kirchhoff Institute for Physics, Heidelberg University, Heidelberg, 69120, Germany.

European Institute for Neuromorphic Computing, Heidelberg University, Heidelberg, 69120, Germany.

出版信息

Open Res Eur. 2024 Nov 14;3:144. doi: 10.12688/openreseurope.15775.2. eCollection 2023.

DOI:10.12688/openreseurope.15775.2
PMID:39669416
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC11635192/
Abstract

BACKGROUND

Finding appropriate model parameters for multi-compartmental neuron models can be challenging. Parameters such as the leak and axial conductance are not always directly derivable from neuron observations but are crucial for replicating desired observations. The objective of this study is to replicate the attenuation behavior of an excitatory postsynaptic potential (EPSP) traveling along a linear chain of compartments on the analog BrainScaleS-2 neuromorphic hardware platform.

METHODS

In the present publication we use genetic algorithms to find suitable model parameters. They promise parameterization without domain knowledge of the neuromorphic substrate or underlying neuron model. To validate the results of the genetic algorithms, a comprehensive grid search was conducted. Furthermore, trial-to-trial variations in the analog system are counteracted utilizing spike-triggered averaging.

RESULTS AND CONCLUSIONS

The algorithm successfully replicated the desired EPSP attenuation behavior in both single and multi-objective searches illustrating the applicability of genetic algorithms to parameterize analog neuromorphic hardware.

摘要

背景

为多房室神经元模型找到合适的模型参数可能具有挑战性。诸如泄漏电导和轴向电导等参数并非总是能直接从神经元观测中得出，但对于复制所需观测结果至关重要。本研究的目的是在模拟BrainScaleS-2神经形态硬件平台上复制沿着线性房室链传播的兴奋性突触后电位（EPSP）的衰减行为。

方法

在本出版物中，我们使用遗传算法来寻找合适的模型参数。它们有望在无需神经形态基质或基础神经元模型领域知识的情况下进行参数化。为了验证遗传算法的结果，进行了全面的网格搜索。此外，利用尖峰触发平均来抵消模拟系统中试验间的变化。

结果与结论

该算法在单目标和多目标搜索中均成功复制了所需的EPSP衰减行为，说明了遗传算法在模拟神经形态硬件参数化方面的适用性。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/a291447ecf43/openreseurope-3-20368-g0009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/3c38b76a13de/openreseurope-3-20368-g0000.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/f4c721272b10/openreseurope-3-20368-g0001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/2416d857bb1f/openreseurope-3-20368-g0002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/93ff06c75880/openreseurope-3-20368-g0003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/00ce9db35ec9/openreseurope-3-20368-g0004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/f011e71f9abd/openreseurope-3-20368-g0005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/3bc8f3b8c78e/openreseurope-3-20368-g0006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/d1c7a67c6a54/openreseurope-3-20368-g0007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/881951004480/openreseurope-3-20368-g0008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/a291447ecf43/openreseurope-3-20368-g0009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/3c38b76a13de/openreseurope-3-20368-g0000.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/f4c721272b10/openreseurope-3-20368-g0001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/2416d857bb1f/openreseurope-3-20368-g0002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/93ff06c75880/openreseurope-3-20368-g0003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/00ce9db35ec9/openreseurope-3-20368-g0004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/f011e71f9abd/openreseurope-3-20368-g0005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/3bc8f3b8c78e/openreseurope-3-20368-g0006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/d1c7a67c6a54/openreseurope-3-20368-g0007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/881951004480/openreseurope-3-20368-g0008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/0d50/11635348/a291447ecf43/openreseurope-3-20368-g0009.jpg

相似文献

1
Parametrizing analog multi-compartment neurons with genetic algorithms.使用遗传算法对模拟多室神经元进行参数化。
Open Res Eur. 2024 Nov 14;3:144. doi: 10.12688/openreseurope.15775.2. eCollection 2023.
2
Emulating Dendritic Computing Paradigms on Analog Neuromorphic Hardware.在模拟神经形态硬件上模拟树突计算范式。
Neuroscience. 2022 May 1;489:290-300. doi: 10.1016/j.neuroscience.2021.08.013. Epub 2021 Aug 21.
3
Surrogate gradients for analog neuromorphic computing.模拟神经形态计算的替代梯度。
Proc Natl Acad Sci U S A. 2022 Jan 25;119(4). doi: 10.1073/pnas.2109194119.
4
A Scalable Approach to Modeling on Accelerated Neuromorphic Hardware.一种在加速神经形态硬件上进行建模的可扩展方法。
Front Neurosci. 2022 May 18;16:884128. doi: 10.3389/fnins.2022.884128. eCollection 2022.
5
Beyond LIF Neurons on Neuromorphic Hardware.超越神经形态硬件上的LIF神经元。
Front Neurosci. 2022 Jul 5;16:881598. doi: 10.3389/fnins.2022.881598. eCollection 2022.
6
Structural plasticity on an accelerated analog neuromorphic hardware system.在加速模拟神经形态硬件系统上实现结构可塑性。
Neural Netw. 2021 Jan;133:11-20. doi: 10.1016/j.neunet.2020.09.024. Epub 2020 Oct 12.
7
Scalable network emulation on analog neuromorphic hardware.模拟神经形态硬件上的可扩展网络仿真。
Front Neurosci. 2025 Feb 5;18:1523331. doi: 10.3389/fnins.2024.1523331. eCollection 2024.
8
Mapping Generative Models onto a Network of Digital Spiking Neurons.将生成模型映射到数字尖峰神经元网络上。
IEEE Trans Biomed Circuits Syst. 2016 Aug;10(4):837-54. doi: 10.1109/TBCAS.2016.2539352. Epub 2016 May 18.
9
Neuromorphic hardware databases for exploring structure-function relationships in the brain.用于探索大脑结构-功能关系的神经形态硬件数据库。
Philos Trans R Soc Lond B Biol Sci. 2001 Aug 29;356(1412):1249-58. doi: 10.1098/rstb.2001.0904.
10
A Conductance-Based Silicon Synapse Circuit.一种基于电导的硅突触电路。
Biomimetics (Basel). 2022 Dec 16;7(4):246. doi: 10.3390/biomimetics7040246.

本文引用的文献

1
Opportunities for neuromorphic computing algorithms and applications.神经形态计算算法与应用的机遇。
Nat Comput Sci. 2022 Jan;2(1):10-19. doi: 10.1038/s43588-021-00184-y. Epub 2022 Jan 31.
2
A Scalable Approach to Modeling on Accelerated Neuromorphic Hardware.一种在加速神经形态硬件上进行建模的可扩展方法。
Front Neurosci. 2022 May 18;16:884128. doi: 10.3389/fnins.2022.884128. eCollection 2022.
3
The BrainScaleS-2 Accelerated Neuromorphic System With Hybrid Plasticity.具有混合可塑性的BrainScaleS-2加速神经形态系统
Front Neurosci. 2022 Feb 24;16:795876. doi: 10.3389/fnins.2022.795876. eCollection 2022.
4
Emulating Dendritic Computing Paradigms on Analog Neuromorphic Hardware.在模拟神经形态硬件上模拟树突计算范式。
Neuroscience. 2022 May 1;489:290-300. doi: 10.1016/j.neuroscience.2021.08.013. Epub 2021 Aug 21.
5
Dendritic action potentials and computation in human layer 2/3 cortical neurons.人类皮层 2/3 层神经元的树突动作电位和计算。
Science. 2020 Jan 3;367(6473):83-87. doi: 10.1126/science.aax6239.
6
Dendritic solutions to the credit assignment problem.树突状解决信用分配问题。
Curr Opin Neurobiol. 2019 Feb;54:28-36. doi: 10.1016/j.conb.2018.08.003. Epub 2018 Sep 8.
7
Systematic generation of biophysically detailed models for diverse cortical neuron types.为多种皮层神经元类型系统地生成生物物理细节模型。
Nat Commun. 2018 Feb 19;9(1):710. doi: 10.1038/s41467-017-02718-3.
8
Training and operation of an integrated neuromorphic network based on metal-oxide memristors.基于金属氧化物忆阻器的集成神经形态网络的训练和操作。
Nature. 2015 May 7;521(7550):61-4. doi: 10.1038/nature14441.
9
Learning by the dendritic prediction of somatic spiking.通过树突预测躯体发放进行学习。
Neuron. 2014 Feb 5;81(3):521-8. doi: 10.1016/j.neuron.2013.11.030.
10
PyNN: A Common Interface for Neuronal Network Simulators.PyNN：神经元网络模拟器的通用接口。
Front Neuroinform. 2009 Jan 27;2:11. doi: 10.3389/neuro.11.011.2008. eCollection 2008.