脑机接口 / BCI - arXivDaily 专题

1810.09391 2018-10-23 cs.LG cs.AI q-bio.NC stat.ML 62%

A neuro-inspired architecture for unsupervised continual learning based on online clustering and hierarchical predictive coding

Constantine Dovrolis

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

Comments Under peer-review

1810.02584 2018-10-08 eess.SP q-bio.NC 62%

Deep Learning for micro-Electrocorticographic (μECoG) Data

Xi Wang, C. Alexis Gkogkidis, Robin T. Schirrmeister, Felix A. Heilmeyer, Mortimer Gierthmuehlen, Fabian Kohler, Martin Schuettler, Thomas Stieglitz, Tonio Ball

专题命中神经信号处理：EEG(abstract)；分类 q-bio.NC、eess.SP

Comments 6 pages, 7 figures, 2018 IEEE EMBS conference

1803.06622 2018-08-21 q-bio.NC cs.AI cs.LG cs.NE 62%

Learning recurrent dynamics in spiking networks

Christopher Kim, Carson Chow

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

1805.10734 2018-05-31 q-bio.NC cs.CV cs.LG 62%

A neural network trained to predict future video frames mimics critical properties of biological neuronal responses and perception

William Lotter, Gabriel Kreiman, David Cox

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

1712.04602 2018-05-17 q-bio.NC cs.IT cs.LG cs.NE math.IT 62%

On the organization of grid and place cells: Neural de-noising via subspace learning

David M. Schwartz, O. Ozan Koyluoglu

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

1706.01380 2017-08-08 q-bio.NC cs.CV cs.LG 62%

Automatic Response Assessment in Regions of Language Cortex in Epilepsy Patients Using ECoG-based Functional Mapping and Machine Learning

Harish RaviPrakash, Milena Korostenskaja, Eduardo Castillo, Ki Lee, James Baumgartner, Ulas Bagci

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

Comments This paper will appear in the Proceedings of IEEE International Conference on Systems, Man and Cybernetics (SMC) 2017

1706.07555 2017-06-27 q-bio.NC cs.LG 62%

Toward Goal-Driven Neural Network Models for the Rodent Whisker-Trigeminal System

Chengxu Zhuang, Jonas Kubilius, Mitra Hartmann, Daniel Yamins

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

Comments 17 pages including supplementary information, 8 figures

1607.00455 2016-11-15 cs.LG q-bio.NC stat.ML 62%

Alzheimer's Disease Diagnostics by Adaptation of 3D Convolutional Network

Ehsan Hosseini-Asl, Robert Keynto, Ayman El-Baz

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

Comments This paper is accepted for publication at IEEE ICIP 2016 conference

1607.00556 2016-07-05 cs.LG q-bio.NC stat.ML 62%

Alzheimer's Disease Diagnostics by a Deeply Supervised Adaptable 3D Convolutional Network

Ehsan Hosseini-Asl, Georgy Gimel'farb, Ayman El-Baz

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

1312.6115 2014-03-25 stat.ML cs.LG cs.NE q-bio.NC 62%

Neuronal Synchrony in Complex-Valued Deep Networks

David P. Reichert, Thomas Serre

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

Comments ICLR 2014, accepted to conference track. This version: added proceedings note, minor additions

1312.6108 2014-02-18 cs.NE cs.LG q-bio.NC 62%

Modeling correlations in spontaneous activity of visual cortex with centered Gaussian-binary deep Boltzmann machines

Nan Wang, Dirk Jancke, Laurenz Wiskott

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

Comments 9 pages, 4 figures, for openreview ICLR2014, 2nd revision

1202.4482 2013-02-12 q-bio.NC cs.LG nlin.AO 62%

Metabolic cost as an organizing principle for cooperative learning

David Balduzzi, Pedro A Ortega, Michel Besserve

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

Comments 14 pages, 2 figures, to appear in Advances in Complex Systems

1211.1255 2012-11-07 cs.LG cs.CV q-bio.NC 62%

Handwritten digit recognition by bio-inspired hierarchical networks

Antonio G. Zippo, Giuliana Gelsomino, Sara Nencini, Gabriele E. M. Biella

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

1108.2840 2011-08-16 q-bio.NC cs.LG stat.ML 62%

Generalised elastic nets

Miguel Á. Carreira-Perpiñán, Geoffrey J. Goodhill

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC、cs.LG

Comments 52 pages, 16 figures. Original manuscript dated August 14, 2003 and not updated since. Current authors' email addresses: mcarreira-perpinan@ucmerced.edu, g.goodhill@uq.edu.au

2205.07045 2022-05-17 q-bio.NC q-bio.QM 61%

How can I investigate causal brain networks with iEEG?

Yuhao Huang, Corey Keller

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC；EEG(comments)

Comments Forthcoming chapter in "Intracranial EEG for Cognitive Neuroscience"

2608.12377 2026-08-14 q-bio.NC cs.AI cs.CL 新提交 57%

From Observation to Intervention: Memory in Brains and Large Language Models

从观察到干预：大脑与大型语言模型中的记忆

Morteza Salehjahromi, Shayan A. Zadegan, Amgad Muneer, Jia Wu

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC

AI总结该研究对比大脑与LLMs的记忆系统，提出可利用LLMs在实验访问上的优势，将记忆相关功能问题转化为更精准的生物学假设，核心是转移实验逻辑而非解剖结构。

Comments Perspective article, 11 pages, 3 figures, 1 table, and 1 key terms box. Submitted for consideration to Nature Machine Intelligence

2608.12717 2026-08-14 cs.LG cs.CL 新提交 57%

Perturbation-based Regional Interpretability through Subtraction Mapping (PRISM): naming-error dissociations in language models and post-stroke aphasia

基于减法映射的扰动型区域可解释性方法（PRISM）：语言模型与卒中后失语症中的命名错误分离现象

Xiang Guan, Roger D. Newman-Norlund, Yong Yang, Saeed Ahmadi, Regan Willis, Nadra Salman, Kalil Warren, Srihari Nelakuditi, Chris Rorden, Leonardo Bonilha, Julius Fridriksson

机构 * University of South Carolina（南卡罗来纳大学）； ALLT.AI, LLC（ALLT.AI有限责任公司）； USC School of Medicine（南卡罗来纳大学医学院）

专题命中神经信号处理：cortical(abstract)；分类 cs.LG

AI总结本研究开发PRISM方法，将神经成像减法分析适配至Transformer语言模型，对比模型与213名卒中后失语症患者的命名错误模式，验证了音素偏向性分离等结果，为模型功能专业化提供可证伪的空间分辨率测试工具。

Comments 49 pages, 6 figures, 1 table. Supplementary methods, 6 tables and 5 figures included

2608.10211 2026-08-12 q-bio.NC 新提交 57%

Reduced Gibbs free energy supply hinders brain information processing during mental fatigue

吉布斯自由能供应减少阻碍精神疲劳期间的大脑信息处理

Danko D. Georgiev, Oskan B. Tasinov, Danail V. Pavlov, Deyan S. Hrusafov

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC

AI总结该研究通过计算建模发现，精神疲劳时ATP水解供能减少会损害蛋白质α-螺旋的酰胺I激子协同效应，改变神经元电生理特性，削弱大脑信息处理能力，建议智力工作中间歇性恢复以保护健康和生产力。

Comments 21 pages, 4 figures, 1 table

Journal ref Fatigue: Biomedicine, Health & Behavior 2026; 14 (3): 216-233

2608.08317 2026-08-11 cs.LG 新提交 57%

The Neural Division of Labor: Biologically-Inspired Modular Architectures for Robust Neuromorphic Computing

神经劳动分工：用于鲁棒神经形态计算的生物启发式模块化架构

Maksim Bazhenov, Serafim Grubas, Vakhtang Putkaradze

机构 * University of California San Diego（加州大学圣地亚哥分校）； University of Alberta（阿尔伯塔大学）； University of Alabama（阿拉巴马大学）

专题命中神经信号处理：neural signal(abstract)；分类 cs.LG

AI总结本文提出可分解脉冲神经网络（D-SNN），采用生物启发模块化架构，在减少参数与突触操作的同时，实现鲁棒性与决策透明性，为边缘神经形态计算提供高效方案。

Comments 39 pages, 7 figures

2608.09773 2026-08-11 q-bio.NC physics.comp-ph 新提交 57%

Graph Analysis of Neuronal-Culture Connectivity Derived from a Reservoir-Computing Model

基于储备池计算模型的神经元培养物连接性的图分析

Ilya Auslender, Giorgio Letti, Yasaman Heydari, Lorenzo Pavesi

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC

AI总结本研究基于储备池计算（RC）框架从多通道电生理记录提取神经元培养物的内在连接图谱（ICM），通过图论分析验证了RC连接性推断的有效性，建立了神经元培养物功能网络表征的可扩展数据驱动框架。

2509.24039 2026-08-11 q-bio.NC cs.AI cs.CV cs.SD 版本更新 57%

Topographic Constraints Shape Brain-Like Component Structure in Auditory Models

地形约束塑造听觉模型中类脑的组件结构

Haider Al-Tahan, Mayukh Deb, Jenelle Feather, N. Apurva Ratan Murty

机构 * Cognition and Brain Science, School of Psychology, Georgia Tech（认知与脑科学系，心理学学院，佐治亚理工学院）； Computational Cognition, Georgia Tech（计算认知学，佐治亚理工学院）； Neuroscience Institute and Department of Psychology, Carnegie Mellon University（神经科学研究所和心理学系，卡内基梅隆大学）

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC

AI总结该研究探究地形约束对听觉神经网络模型的影响，提出TopoAudio模型，发现其性能与标准模型相当且内部表示更紧凑、组件与人类ECoG记录对齐更紧密，为人工神经网络生成生物对齐表示提供了机制证据。

2608.05481 2026-08-07 q-bio.NC 新提交 57%

From Local Learning to Global Prediction Through Layered Surprise Cascades

通过分层惊喜级联从局部学习到全局预测

Andrew L. Smith, Linxing Preston Jiang, Jason K. Eshraghian, Matthew S. Bull, Stefano Recanatesi

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC

AI总结该研究提出前向-前向算法的循环变体，从局部学习规则产生预测编码关键原则，为神经科学与机器学习搭建新桥梁。

2608.05462 2026-08-07 physics.bio-ph q-bio.NC 新提交 57%

Toward a Dynamical Taxonomy of Insomnia: A Multiaxial Framework for Sleep-State Transitions and Architectural Failure

迈向失眠的动态分类学：睡眠状态转换与结构异常的多轴框架

Alexander Poltorak

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC

AI总结本研究提出多轴动态框架，以失效的动态运作等为候选特征，用于失眠的分类，其可改善交流、分层与预测，框架为现象学组织模型而非新诊断系统。

Comments 28 pages, 3 figures

2502.14671 2026-08-07 cs.CL cs.AI q-bio.NC 版本更新 57%

Explanations of Large Language Models Explain Language Representations in the Brain

大语言模型的解释可解释大脑中的语言表征

Maryam Rahimi, Mohammad Reza Daliri, Yadollah Yaghoobzadeh

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC

AI总结该研究探究LLM与脑语言表征对齐的驱动因素，利用XAI的归因方法结合fMRI数据，发现梯度归因可稳健对齐脑活动，且不同层归因的特性与脑区域对齐模式存在差异。

2608.00306 2026-08-06 q-bio.NC 版本更新 57%

Mechanistic bridges from receptors to whole-brain dynamics: mean-field reductions, validity domains, and computational trade-offs

从受体到全脑动力学的机制桥梁：平均场约化、有效性域与计算权衡

Yannael Bossard, Lehna Bekri, Alain Destexhe

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC

AI总结本综述梳理受体感知全脑模型的数学谱系，明确其约化假设，将其置于约化群体模型格局中，引入模拟成本与内存流量作为基准维度，界定该建模为多属性间的权衡。

Comments Minor typographical corrections and consistency fixes; conclusions unchanged

2608.03181 2026-08-05 q-bio.NC q-bio.QM 新提交 57%

Persistent homology broadens the controllable subspace in human structural connectomes

持续同调拓宽人类结构连接组中的可控子空间

Carter Sale, Marco Coraggio, Mengsen Zhang, Michael J. Richardson

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC

AI总结该研究提出基于持续拓扑环的驱动节点选择标准，与基于度的标准控制能相近但可控子空间几何更优，能捕捉标量能量遗漏的脑网络控制信息。

Comments 21 pages, 7 figures

2608.01947 2026-08-04 q-bio.NC cs.AI cs.NE 新提交 57%

Divisive Normalization Shapes Low-Rank Slow Manifolds for Continuous Working Memory

归一化分裂塑造连续工作记忆的低秩慢流形

Zhaotian Gu, Jie Su, Weiwei Wang, Chang Liu, Tianyi Qian, Dahui Wang

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC

AI总结本文提出循环归一化分裂网络（RDNN），通过动力学分析和BPTT梯度分析，证明归一化分裂可使网络形成稳健低秩慢流形，防止时变输入下的流形破碎，是学习高保真连续表示的关键机制。

2608.00492 2026-08-04 cs.AI cs.LG 新提交 57%

The Bayesian Reflex: A Predictive Coding Engine for Artificial Intelligence

贝叶斯反射：一种人工智能的预测编码引擎

Sourabh Bhattacharya

机构 * Indian Statistical Institute（印度统计研究所）

专题命中神经信号处理：cortical(abstract)；分类 cs.LG

AI总结该研究提出贝叶斯反射计算框架，结合预测编码三大支柱，利用近期算法突破实现可扩展的类脑持续学习、感知与决策，通过多领域应用为自适应AI提供蓝图。

Comments From the Bayesian Brain to Artificial Intelligence

2607.29068 2026-08-03 q-bio.NC 新提交 57%

Critical Flicker Fusion Frequency As A Falsifiable Boundary Between Plastic And Non-Plastic Neural Systems

临界闪烁融合频率作为可证伪的可塑性与非可塑性神经系统之间的边界

Natalia D. Rydzenska, Pawel J. Winklewski, Michal W. Blaszczyk-Niezgoda, Anna B. Marcinkowska

专题命中神经信号处理：cortical(abstract)；分类 q-bio.NC

AI总结本综述提出临界闪烁融合频率（CFFF）可作为可塑性与非可塑性神经系统的可证伪边界，分析其稳定性来源、对不同训练的反应及相关原则，为理解认知系统在感觉约束下的运作提供新视角。

Comments 44 pages, 4 figures, 3 tables in Supplementary Material

2607.28663 2026-08-03 cs.NE cs.LG 新提交 57%

NeuroSynth: A Biologically Inspired Continual Reinforcement Learning Architecture for Mitigating Catastrophic Forgetting

NeuroSynth：一种受生物启发的持续强化学习架构，用于缓解灾难性遗忘

Yash Kini

专题命中神经信号处理：cortical(abstract)；分类 cs.LG

AI总结本研究提出受生物启发的持续强化学习架构NeuroSynth，通过双路径巩固机制缓解灾难性遗忘，实验显示其在多个顺序导航任务中相比PPO、EWC保留更多早期任务知识，改善了持续学习的稳定性-可塑性平衡。

Comments Disclaimer. This manuscript is provided as an arXiv preprint to establish a public record of the NeuroSynth continual reinforcement learning architecture and its evaluation on the NeuroMaze-CL benchmark. This full manuscript has been submitted to the Journal of High School Science for peer review

1. 神经信号处理 2241 篇

A neuro-inspired architecture for unsupervised continual learning based on online clustering and hierarchical predictive coding

Deep Learning for micro-Electrocorticographic (μECoG) Data

Learning recurrent dynamics in spiking networks

A neural network trained to predict future video frames mimics critical properties of biological neuronal responses and perception

On the organization of grid and place cells: Neural de-noising via subspace learning

Automatic Response Assessment in Regions of Language Cortex in Epilepsy Patients Using ECoG-based Functional Mapping and Machine Learning

Toward Goal-Driven Neural Network Models for the Rodent Whisker-Trigeminal System

Alzheimer's Disease Diagnostics by Adaptation of 3D Convolutional Network

Alzheimer's Disease Diagnostics by a Deeply Supervised Adaptable 3D Convolutional Network

Neuronal Synchrony in Complex-Valued Deep Networks

Modeling correlations in spontaneous activity of visual cortex with centered Gaussian-binary deep Boltzmann machines

Metabolic cost as an organizing principle for cooperative learning

Handwritten digit recognition by bio-inspired hierarchical networks

Generalised elastic nets

How can I investigate causal brain networks with iEEG?

From Observation to Intervention: Memory in Brains and Large Language Models

Perturbation-based Regional Interpretability through Subtraction Mapping (PRISM): naming-error dissociations in language models and post-stroke aphasia

Reduced Gibbs free energy supply hinders brain information processing during mental fatigue

The Neural Division of Labor: Biologically-Inspired Modular Architectures for Robust Neuromorphic Computing

Graph Analysis of Neuronal-Culture Connectivity Derived from a Reservoir-Computing Model

Topographic Constraints Shape Brain-Like Component Structure in Auditory Models

From Local Learning to Global Prediction Through Layered Surprise Cascades

Toward a Dynamical Taxonomy of Insomnia: A Multiaxial Framework for Sleep-State Transitions and Architectural Failure

Explanations of Large Language Models Explain Language Representations in the Brain

Mechanistic bridges from receptors to whole-brain dynamics: mean-field reductions, validity domains, and computational trade-offs

Persistent homology broadens the controllable subspace in human structural connectomes

Divisive Normalization Shapes Low-Rank Slow Manifolds for Continuous Working Memory

The Bayesian Reflex: A Predictive Coding Engine for Artificial Intelligence

Critical Flicker Fusion Frequency As A Falsifiable Boundary Between Plastic And Non-Plastic Neural Systems

NeuroSynth: A Biologically Inspired Continual Reinforcement Learning Architecture for Mitigating Catastrophic Forgetting