Developing novel neural network tools for accurate and interpretable dynamical modeling of neural circuits

开发新型神经网络工具,用于准确且可解释的神经回路动态建模

基本信息

  • 批准号:
    10752956
  • 负责人:
  • 金额:
    $ 7.66万
  • 依托单位:
  • 依托单位国家:
    美国
  • 项目类别:
  • 财政年份:
    2023
  • 资助国家:
    美国
  • 起止时间:
    2023-08-01 至 2026-07-31
  • 项目状态:
    未结题

项目摘要

Abstract In recent years, the number of neurons that we can record simultaneously has seen an exponential increase, presenting a daunting challenge: how do we analyze these complex and high-dimensional datasets to gain insight into how neural circuits perform computation? Tools from dynamical systems theory have successfully unraveled the computational machinery of artificial recurrent neural networks (RNNs) trained to perform goal-directed tasks. If we could apply these tools to biological neural circuits, it would provide unparalleled access to the inner workings of the brain and potentially allow us to connect theories of neural computation to real biological data. However, for these tools to be useful, we need to create in silico replicas whose dynamics faithfully represent the dynamics of the underlying biological system. To date, the best in silico replicas of biological networks are RNNs trained to produce output that matches recorded patterns of neuronal firing. While this approach is rapidly growing in popularity, it has critical flaws. Current training methodologies are not constrained to produce accurate representations of the underlying dynamics; in fact, RNNs are actually rewarded for inventing superfluous dynamics, so long as those dynamics help to reproduce recorded neural data. Additionally, these models often assume that the relationship (“embedding”) between latent activity and neural firing rates is linear; when this assumption proves false, the dynamical accuracy suffers. The problems of superfluous dynamics and non-linear embedding are especially severe when attempting to model a system of interacting neural circuits. The objective of this proposal is to develop a novel artificial neural network architecture that addresses the above challenges and allows our in-silico models to capture accurate dynamics that are built both within and across-circuits. My approach combines two key components: 1) neural ordinary differential equations (NODEs), a computational architecture that we have demonstrated learns dynamics more accurately and compactly than RNNs and 2) invertible neural network (INN) readouts, which eliminate superfluous dynamics and allow the model to approximate nonlinear embeddings. I will validate the ability of this model, called an Ordinary Differential equation auto-encoder with Invertible readout (ODIN), to find accurate within- and across-circuit dynamics using synthetic neural data and previously-collected multi-electrode recordings from monkeys. This tool will help to build a bridge between neural data and both local and distributed neural computations.
摘要 近年来,我们可以同时记录的神经元数量呈指数级增长, 增加,提出了一个艰巨的挑战:我们如何分析这些复杂的和高维的 数据集来深入了解神经回路如何执行计算?动力系统工具 理论已经成功地解开了人工递归神经网络的计算机制 (RNN)训练执行目标导向的任务。如果我们能将这些工具应用于生物神经回路, 它将提供前所未有的进入大脑内部工作的途径,并可能使我们能够 将神经计算理论与真实的生物数据联系起来。然而,为了使这些工具有用, 我们需要在计算机上创建复制品,其动态忠实地代表底层的动态, 生物系统。 到目前为止,生物网络的最佳计算机复制品是经过训练的RNN,其输出 与记录的神经元放电模式相匹配虽然这种方法正在迅速普及,但它 关键缺陷目前的训练方法并不局限于产生准确的表示 实际上,RNN实际上是因为发明了多余的动力学而受到奖励, 只要这些动力学有助于再现记录的神经数据。此外,这些模型通常 假设潜在活动和神经放电率之间的关系(“嵌入”)是线性的;当 这个假设被证明是错误的,动态精度受到影响。多余动力学问题 和非线性嵌入是特别严重时,试图模拟系统的相互作用 神经回路 该建议的目的是开发一种新的人工神经网络架构, 解决了上述挑战,并允许我们的计算机模型捕捉准确的动态, 内置和跨电路。我的方法结合了两个关键组成部分:1)神经普通 微分方程(NODE),我们已经证明学习的计算架构 动态比RNN更精确和更复杂,以及2)可逆神经网络(INN)读数, 这消除了多余的动态并允许模型近似非线性嵌入。我会 验证了该模型的能力,称为常微分方程自动编码器与可逆 读出(ODIN),使用合成神经数据找到准确的内部和跨电路动态, 先前从猴子身上收集的多电极记录。这个工具将有助于建立一个桥梁 神经数据与本地和分布式神经计算之间的关系。

项目成果

期刊论文数量(0)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)

数据更新时间:{{ journalArticles.updateTime }}

{{ item.title }}
{{ item.translation_title }}
  • DOI:
    {{ item.doi }}
  • 发表时间:
    {{ item.publish_year }}
  • 期刊:
  • 影响因子:
    {{ item.factor }}
  • 作者:
    {{ item.authors }}
  • 通讯作者:
    {{ item.author }}

数据更新时间:{{ journalArticles.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ monograph.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ sciAawards.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ conferencePapers.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ patent.updateTime }}

Christopher Versteeg其他文献

Christopher Versteeg的其他文献

{{ item.title }}
{{ item.translation_title }}
  • DOI:
    {{ item.doi }}
  • 发表时间:
    {{ item.publish_year }}
  • 期刊:
  • 影响因子:
    {{ item.factor }}
  • 作者:
    {{ item.authors }}
  • 通讯作者:
    {{ item.author }}

{{ truncateString('Christopher Versteeg', 18)}}的其他基金

Determining the role of the Cuneate nucleus in the processing of proprioceptive information in the awake behaving animal
确定楔形核在清醒行为动物本体感觉信息处理中的作用
  • 批准号:
    9812769
  • 财政年份:
    2018
  • 资助金额:
    $ 7.66万
  • 项目类别:

相似海外基金

RII Track-4:NSF: From the Ground Up to the Air Above Coastal Dunes: How Groundwater and Evaporation Affect the Mechanism of Wind Erosion
RII Track-4:NSF:从地面到沿海沙丘上方的空气:地下水和蒸发如何影响风蚀机制
  • 批准号:
    2327346
  • 财政年份:
    2024
  • 资助金额:
    $ 7.66万
  • 项目类别:
    Standard Grant
BRC-BIO: Establishing Astrangia poculata as a study system to understand how multi-partner symbiotic interactions affect pathogen response in cnidarians
BRC-BIO:建立 Astrangia poculata 作为研究系统,以了解多伙伴共生相互作用如何影响刺胞动物的病原体反应
  • 批准号:
    2312555
  • 财政年份:
    2024
  • 资助金额:
    $ 7.66万
  • 项目类别:
    Standard Grant
How Does Particle Material Properties Insoluble and Partially Soluble Affect Sensory Perception Of Fat based Products
不溶性和部分可溶的颗粒材料特性如何影响脂肪基产品的感官知觉
  • 批准号:
    BB/Z514391/1
  • 财政年份:
    2024
  • 资助金额:
    $ 7.66万
  • 项目类别:
    Training Grant
Graduating in Austerity: Do Welfare Cuts Affect the Career Path of University Students?
紧缩毕业:福利削减会影响大学生的职业道路吗?
  • 批准号:
    ES/Z502595/1
  • 财政年份:
    2024
  • 资助金额:
    $ 7.66万
  • 项目类别:
    Fellowship
Insecure lives and the policy disconnect: How multiple insecurities affect Levelling Up and what joined-up policy can do to help
不安全的生活和政策脱节:多种不安全因素如何影响升级以及联合政策可以提供哪些帮助
  • 批准号:
    ES/Z000149/1
  • 财政年份:
    2024
  • 资助金额:
    $ 7.66万
  • 项目类别:
    Research Grant
感性個人差指標 Affect-X の構築とビスポークAIサービスの基盤確立
建立个人敏感度指数 Affect-X 并为定制人工智能服务奠定基础
  • 批准号:
    23K24936
  • 财政年份:
    2024
  • 资助金额:
    $ 7.66万
  • 项目类别:
    Grant-in-Aid for Scientific Research (B)
How does metal binding affect the function of proteins targeted by a devastating pathogen of cereal crops?
金属结合如何影响谷类作物毁灭性病原体靶向的蛋白质的功能?
  • 批准号:
    2901648
  • 财政年份:
    2024
  • 资助金额:
    $ 7.66万
  • 项目类别:
    Studentship
ERI: Developing a Trust-supporting Design Framework with Affect for Human-AI Collaboration
ERI:开发一个支持信任的设计框架,影响人类与人工智能的协作
  • 批准号:
    2301846
  • 财政年份:
    2023
  • 资助金额:
    $ 7.66万
  • 项目类别:
    Standard Grant
Investigating how double-negative T cells affect anti-leukemic and GvHD-inducing activities of conventional T cells
研究双阴性 T 细胞如何影响传统 T 细胞的抗白血病和 GvHD 诱导活性
  • 批准号:
    488039
  • 财政年份:
    2023
  • 资助金额:
    $ 7.66万
  • 项目类别:
    Operating Grants
How motor impairments due to neurodegenerative diseases affect masticatory movements
神经退行性疾病引起的运动障碍如何影响咀嚼运动
  • 批准号:
    23K16076
  • 财政年份:
    2023
  • 资助金额:
    $ 7.66万
  • 项目类别:
    Grant-in-Aid for Early-Career Scientists
{{ showInfoDetail.title }}

作者:{{ showInfoDetail.author }}

知道了