Physica A: Statistical Mechanics and its Applications最新文献_第8页

In Memoriam J. Robert “Bob” Dorfman (1937–2025) j·罗伯特·“鲍勃”·多尔夫曼（1937-2025）

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2025-12-26 DOI: 10.1016/j.physa.2025.131247

T.R. Kirkpatrick , J.V. Sengers , H. van Beijeren

引用次数: 0

Energy consumption of spontaneous up–down oscillations modulated by ion channel inactivation 离子通道失活调制的自发上下振荡的能量消耗

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2025-12-26 DOI: 10.1016/j.physa.2025.131242

Xiaoqian Liu , Lulu Lu , Yuan Zhu , Ming Yi

In clinical practice, sodium and potassium channels are considered targets of action for a variety of drugs (e.g., propofol, curcumin, etc.). Most of the studies are based on physiological experiments in which researchers have investigated the effects of ion channel residues on the conductance and homeostatic inactivation processes of ion channels by means of gene targeting mutation techniques. However, few researchers investigate the effects of voltage-gated ion channel properties on neural network behavior based on kinetic models. In this paper, we systematically vary the conductivity per unit area or the inactivation time constant to model the effects of drugs on ion channels. Numerical results show that, persistent sodium dynamics affect the firing rate and energy consumption of the neural network as much as fast sodium dynamics does. And the high synchronization rate of the neural network significantly enhances the firing rate of the network. In addition, we find that the network’s excitability is significantly increased by enhancing potassium channel inactivation, whereas enhanced inactivation of sodium channels can rapidly inhibit network excitability. These simulations are in general consistent with the experimental results. Our research provides a perspective on understanding the process of drug regulation of ion channels from a kinetic perspective.

在临床实践中，钠和钾通道被认为是多种药物（如异丙酚、姜黄素等）的作用靶点。大多数研究都是基于生理实验，研究人员通过基因靶向突变技术研究了离子通道残基对离子通道电导和稳态失活过程的影响。然而，基于动力学模型研究电压门控离子通道特性对神经网络行为的影响的研究很少。在本文中，我们系统地改变单位面积电导率或失活时间常数来模拟药物对离子通道的影响。数值结果表明，持久的钠离子动力学与快速的钠离子动力学对神经网络发射速率和能量消耗的影响是一样的。神经网络的高同步率显著提高了网络的放电速率。此外，我们发现通过加强钾通道的失活可以显著提高网络的兴奋性，而加强钠通道的失活可以迅速抑制网络的兴奋性。这些模拟结果与实验结果基本一致。我们的研究为从动力学角度理解药物调控离子通道的过程提供了一个视角。

{"title":"Energy consumption of spontaneous up–down oscillations modulated by ion channel inactivation","authors":"Xiaoqian Liu , Lulu Lu , Yuan Zhu , Ming Yi","doi":"10.1016/j.physa.2025.131242","DOIUrl":"10.1016/j.physa.2025.131242","url":null,"abstract":"<div><div>In clinical practice, sodium and potassium channels are considered targets of action for a variety of drugs (e.g., propofol, curcumin, etc.). Most of the studies are based on physiological experiments in which researchers have investigated the effects of ion channel residues on the conductance and homeostatic inactivation processes of ion channels by means of gene targeting mutation techniques. However, few researchers investigate the effects of voltage-gated ion channel properties on neural network behavior based on kinetic models. In this paper, we systematically vary the conductivity per unit area or the inactivation time constant to model the effects of drugs on ion channels. Numerical results show that, persistent sodium dynamics affect the firing rate and energy consumption of the neural network as much as fast sodium dynamics does. And the high synchronization rate of the neural network significantly enhances the firing rate of the network. In addition, we find that the network’s excitability is significantly increased by enhancing potassium channel inactivation, whereas enhanced inactivation of sodium channels can rapidly inhibit network excitability. These simulations are in general consistent with the experimental results. Our research provides a perspective on understanding the process of drug regulation of ion channels from a kinetic perspective.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"684 ","pages":"Article 131242"},"PeriodicalIF":3.1,"publicationDate":"2025-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145886376","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Emergent traffic flow patterns from different intelligent levels of autonomous vehicles in fog 不同智能水平的自动驾驶汽车在雾中的紧急交通流模式

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2025-12-24 DOI: 10.1016/j.physa.2025.131239

Yongsheng Qian, Qingru Zhang, Junwei Zeng, Xu Wei

Before the widespread adoption of L4 and L5 autonomous vehicles, the road system will remain in a phase where L2 and L3 autonomous vehicles mix with human-driven vehicles for a long time. Low visibility in foggy conditions further affects both the traffic efficiency and safety of the mixed traffic flow. This paper considers the differences in driving characteristics of L0, L2, L3, and L4 autonomous vehicles under foggy conditions, constructs a two-lane cellular automata model for foggy highways, and calibrates the parameters of the Gipps model using actual car-following data. Through simulations under varying visibility conditions, we analyze the impact of autonomous vehicles at different levels on overall traffic flow. The results show that the autonomous vehicles can significantly enhance the capacity and average speed of the mixed traffic flows while reducing the congestion rate, and these benefits become more pronounced as the level of automation increases. Moreover, due to varying visibility, road density and penetration rates in different foggy scenarios, there are notable differences in how vehicles of different automation levels affect traffic safety (TET、TERCRI) and stability (SD). Within specific density ranges, the presence of autonomous vehicles improves the safety and stability of mixed traffic flow.

在L4和L5级自动驾驶汽车被广泛采用之前，道路系统将长期处于L2和L3级自动驾驶汽车与人类驾驶汽车混合的阶段。大雾条件下的低能见度进一步影响混合交通流的交通效率和安全。本文考虑雾天条件下L0、L2、L3和L4自动驾驶汽车的驾驶特性差异，构建雾天公路双车道元胞自动机模型，并利用实际车辆跟随数据对Gipps模型参数进行标定。通过不同能见度条件下的仿真，分析了不同水平的自动驾驶汽车对整体交通流的影响。结果表明，自动驾驶汽车可以显著提高混合交通流的通行能力和平均速度，同时降低拥堵率，并且随着自动化水平的提高，这些效益更加明显。此外，由于不同雾天场景下能见度、道路密度和渗透率的不同，不同自动化水平的车辆对交通安全（TET、TERCRI）和稳定性（SD）的影响也存在显著差异。在特定的密度范围内，自动驾驶车辆的存在提高了混合交通流的安全性和稳定性。

{"title":"Emergent traffic flow patterns from different intelligent levels of autonomous vehicles in fog","authors":"Yongsheng Qian, Qingru Zhang, Junwei Zeng, Xu Wei","doi":"10.1016/j.physa.2025.131239","DOIUrl":"10.1016/j.physa.2025.131239","url":null,"abstract":"<div><div>Before the widespread adoption of L4 and L5 autonomous vehicles, the road system will remain in a phase where L2 and L3 autonomous vehicles mix with human-driven vehicles for a long time. Low visibility in foggy conditions further affects both the traffic efficiency and safety of the mixed traffic flow. This paper considers the differences in driving characteristics of L0, L2, L3, and L4 autonomous vehicles under foggy conditions, constructs a two-lane cellular automata model for foggy highways, and calibrates the parameters of the Gipps model using actual car-following data. Through simulations under varying visibility conditions, we analyze the impact of autonomous vehicles at different levels on overall traffic flow. The results show that the autonomous vehicles can significantly enhance the capacity and average speed of the mixed traffic flows while reducing the congestion rate, and these benefits become more pronounced as the level of automation increases. Moreover, due to varying visibility, road density and penetration rates in different foggy scenarios, there are notable differences in how vehicles of different automation levels affect traffic safety (TET、TERCRI) and stability (SD). Within specific density ranges, the presence of autonomous vehicles improves the safety and stability of mixed traffic flow.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"683 ","pages":"Article 131239"},"PeriodicalIF":3.1,"publicationDate":"2025-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145841020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Evacuation performance of pedestrian flow under limited visibility in straight corridor 有限能见度下直道人流疏散性能研究

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2025-12-23 DOI: 10.1016/j.physa.2025.131230

Maoyu Li , Zhizuan Zhou , Dimeng Lai , Lizhong Yang

In the event of power outages or fires, visibility in buildings can be severely reduced, posing serious challenges to safe evacuation. However, the impact of limited visibility on pedestrian flow dynamics remains poorly understood. To address this gap, this study conducted a series of controlled pedestrian flow experiments in a straight corridor, examining the impact of both normal and limited visibility across different crowd densities. Results show that lane formation as a typical self-organizing phenomenon occurred under both visibility conditions. The difference in pedestrian speed between the two visibility levels exhibits a piecewise linear relationship with density: in the relatively free stage (ρ < 2 ped/m²), speed differences decline with increasing density, while in the constrained stage (ρ ≥ 2 ped/m²), the difference stabilizes at a lower level. In addition, under limited visibility, pedestrian flow rate maintains linearly related to exit width but is consistently lower than that under normal visibility. The flow also becomes less continuous, with a higher likelihood of congestion near the exit. From a spatial perspective, limited visibility leads to higher local densities, as evidenced by smaller occupied areas and shorter distances to nearest neighbors, indicating the formation of a more compact crowd structure. These findings enhance our understanding of pedestrian flow under limited visibility and offer valuable insights for the development of evacuation strategies and crowd management. In practical applications, crowd control strategies under limited visibility should be adapted according to the dominant factors at different density levels.

在停电或火灾的情况下，建筑物的能见度会严重降低，给安全疏散带来严重挑战。然而，能见度有限对行人流量动力学的影响仍然知之甚少。为了解决这一问题，本研究在一条笔直的走廊上进行了一系列控制行人流量的实验，研究了正常能见度和有限能见度在不同人群密度下的影响。结果表明，在两种能见度条件下，车道形成都是一种典型的自组织现象。两种能见度水平之间的行人速度差与密度呈分段线性关系：在相对自由阶段（ρ < 2 ped/m2），速度差随密度的增加而减小，而在受限阶段（ρ≥2 ped/m2），速度差稳定在较低水平。此外，在有限能见度下，行人流量与出口宽度保持线性关系，但始终低于正常能见度下的行人流量。车流也变得不那么连续，在出口附近出现拥堵的可能性更高。从空间角度来看，能见度有限导致局部密度较高，被占用的面积更小，与最近邻居的距离更短，表明形成了更紧凑的人群结构。这些发现增强了我们对有限能见度下行人流量的理解，并为疏散策略的制定和人群管理提供了有价值的见解。在实际应用中，在能见度有限的情况下，应根据不同密度下的优势因素，制定相应的人群控制策略。

{"title":"Evacuation performance of pedestrian flow under limited visibility in straight corridor","authors":"Maoyu Li , Zhizuan Zhou , Dimeng Lai , Lizhong Yang","doi":"10.1016/j.physa.2025.131230","DOIUrl":"10.1016/j.physa.2025.131230","url":null,"abstract":"<div><div>In the event of power outages or fires, visibility in buildings can be severely reduced, posing serious challenges to safe evacuation. However, the impact of limited visibility on pedestrian flow dynamics remains poorly understood. To address this gap, this study conducted a series of controlled pedestrian flow experiments in a straight corridor, examining the impact of both normal and limited visibility across different crowd densities. Results show that lane formation as a typical self-organizing phenomenon occurred under both visibility conditions. The difference in pedestrian speed between the two visibility levels exhibits a piecewise linear relationship with density: in the relatively free stage (<em>ρ</em> < 2 ped/m<sup>2</sup>), speed differences decline with increasing density, while in the constrained stage (<em>ρ</em> ≥ 2 ped/m<sup>2</sup><em>)</em>, the difference stabilizes at a lower level. In addition, under limited visibility, pedestrian flow rate maintains linearly related to exit width but is consistently lower than that under normal visibility. The flow also becomes less continuous, with a higher likelihood of congestion near the exit. From a spatial perspective, limited visibility leads to higher local densities, as evidenced by smaller occupied areas and shorter distances to nearest neighbors, indicating the formation of a more compact crowd structure. These findings enhance our understanding of pedestrian flow under limited visibility and offer valuable insights for the development of evacuation strategies and crowd management. In practical applications, crowd control strategies under limited visibility should be adapted according to the dominant factors at different density levels.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"683 ","pages":"Article 131230"},"PeriodicalIF":3.1,"publicationDate":"2025-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145841019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An information propagation model based on rumor-debunking and multi-dimensional influence 基于谣言揭穿和多维影响的信息传播模型

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2025-12-22 DOI: 10.1016/j.physa.2025.131229

Rong Wang , Lingqi Deng , Wenbo Yuan , Yuke Xu , Beilei Wang , Yunpeng Xiao

To address the challenges of data sparsity and the complex interplay of rumor and debunking information in the early stages of rumor propagation, an Information Propagation Model Based on Rumor-Debunking and Multi-Dimensional Influence is proposed. Firstly, to mitigate data sparsity, a three-dimensional tensor model of “active user-potential user-interaction behavior” is constructed. The three-dimensional tensor model leverages the low-rank approximation properties of tensor decomposition to extract hidden relationships and weak correlation patterns between potential and active users, effectively alleviating the issue of sparse rumor data. Secondly, game theory is introduced to quantify the influence of multi-level interactions. A quantification framework is designed for multi-type topic message drivers. This framework considers the competitive relationships among different messages. A dynamic multi-message-oriented interaction strategy is employed to model the influence of multi-source information flow decisions quantitatively. Finally, considering multiple influencing factors in rumor dissemination and the dynamic nature of network topology, a multi-dimensional influence coupling matrix is constructed. This matrix is used to reconstruct the user relationship networks. Additionally, a time-aware graph attention mechanism is introduced to model the sequential dependencies of user interactions. This mechanism leverages spatiotemporal joint embedding to enhance the ability to capture the dynamic evolution of user decision-making. Experimental results demonstrate that the proposed model effectively captures the influence of multi-dimensional factors on user behavior during rumor-debunking topic propagation and significantly enhances the prediction of topic dissemination dynamics. Compared to baseline models, the proposed method improves prediction accuracy by an average of 4.8%.

针对谣言传播初期数据稀疏和谣言与辟谣信息复杂相互作用的挑战，提出了一种基于谣言-辟谣和多维影响的信息传播模型。首先，为了缓解数据稀疏性，构建了“活跃用户-潜在用户交互行为”的三维张量模型；三维张量模型利用张量分解的低秩近似特性，提取潜在用户和活跃用户之间的隐藏关系和弱相关模式，有效缓解谣言数据稀疏的问题。其次，引入博弈论来量化多层次相互作用的影响。设计了多类型主题消息驱动的量化框架。该框架考虑了不同信息之间的竞争关系。采用面向多消息的动态交互策略，对多源信息流决策的影响进行定量建模。最后，考虑谣言传播的多重影响因素和网络拓扑的动态性，构建了多维影响耦合矩阵。该矩阵用于重构用户关系网络。此外，引入了一种时间感知的图注意机制，对用户交互的顺序依赖关系进行建模。该机制利用时空联合嵌入来增强捕获用户决策动态演变的能力。实验结果表明，该模型有效捕捉了辟谣话题传播过程中多维因素对用户行为的影响，显著增强了对话题传播动态的预测能力。与基线模型相比，该方法的预测精度平均提高了4.8%。

{"title":"An information propagation model based on rumor-debunking and multi-dimensional influence","authors":"Rong Wang , Lingqi Deng , Wenbo Yuan , Yuke Xu , Beilei Wang , Yunpeng Xiao","doi":"10.1016/j.physa.2025.131229","DOIUrl":"10.1016/j.physa.2025.131229","url":null,"abstract":"<div><div>To address the challenges of data sparsity and the complex interplay of rumor and debunking information in the early stages of rumor propagation, an Information Propagation Model Based on Rumor-Debunking and Multi-Dimensional Influence is proposed. Firstly, to mitigate data sparsity, a three-dimensional tensor model of “active user-potential user-interaction behavior” is constructed. The three-dimensional tensor model leverages the low-rank approximation properties of tensor decomposition to extract hidden relationships and weak correlation patterns between potential and active users, effectively alleviating the issue of sparse rumor data. Secondly, game theory is introduced to quantify the influence of multi-level interactions. A quantification framework is designed for multi-type topic message drivers. This framework considers the competitive relationships among different messages. A dynamic multi-message-oriented interaction strategy is employed to model the influence of multi-source information flow decisions quantitatively. Finally, considering multiple influencing factors in rumor dissemination and the dynamic nature of network topology, a multi-dimensional influence coupling matrix is constructed. This matrix is used to reconstruct the user relationship networks. Additionally, a time-aware graph attention mechanism is introduced to model the sequential dependencies of user interactions. This mechanism leverages spatiotemporal joint embedding to enhance the ability to capture the dynamic evolution of user decision-making. Experimental results demonstrate that the proposed model effectively captures the influence of multi-dimensional factors on user behavior during rumor-debunking topic propagation and significantly enhances the prediction of topic dissemination dynamics. Compared to baseline models, the proposed method improves prediction accuracy by an average of 4.8%.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"683 ","pages":"Article 131229"},"PeriodicalIF":3.1,"publicationDate":"2025-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145841544","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Posterior collapse as a phase transition in variational autoencoders 变分自编码器中的后向坍缩相变

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2025-12-22 DOI: 10.1016/j.physa.2025.131228

Zhen Li , Fan Zhang , Zheng Zhang , Yu Chen

We investigate the phenomenon of posterior collapse in variational autoencoders (VAEs) from the perspective of statistical physics, and reveal that it constitutes a phase transition governed jointly by data structure and model hyper-parameters. By analyzing the stability of the trivial solution associated with posterior collapse, we identify a critical hyper-parameter threshold. In particular, we derive an explicit criterion for the onset of collapse: posterior collapse occurs when the decoder variance exceeds the largest eigenvalue of the data covariance matrix. This critical boundary, separating meaningful latent inference from collapse, is characterized by a discontinuity in the KL-divergence between the approximate posterior and the prior distribution, where the KL-divergence and its derivatives exhibit clear non-analytic behavior. We validate this critical behavior on both synthetic and real-world datasets, confirming the existence of a phase transition. The experimental results align well with our theoretical predictions, demonstrating the robustness of our collapse criterion across various VAE architectures. Our stability-based analysis demonstrate that posterior collapse is not merely an optimization failure, but rather an emerging phase transition arising from the interplay between data structure and variational constraints. This perspective offers new insights into the trainability and representational capacity of deep generative models.

本文从统计物理的角度研究了变分自编码器（VAEs）的后验坍缩现象，揭示了它是由数据结构和模型超参数共同控制的相变。通过分析与后验崩溃相关的平凡解的稳定性，我们确定了一个临界超参数阈值。特别是，我们推导出一个明确的崩溃开始准则：当解码器方差超过数据协方差矩阵的最大特征值时，后验崩溃发生。这个临界边界将有意义的潜在推断与崩溃分离开来，其特征是近似后验分布和先验分布之间的kl -散度不连续，其中kl -散度及其导数表现出明显的非解析行为。我们在合成和实际数据集上验证了这一关键行为，证实了相变的存在。实验结果与我们的理论预测一致，证明了我们的崩溃准则在各种VAE体系结构中的鲁棒性。我们基于稳定性的分析表明，后验崩溃不仅仅是一种优化失败，而是一种由数据结构和变分约束之间的相互作用引起的新兴相变。这种观点为深度生成模型的可训练性和表征能力提供了新的见解。

{"title":"Posterior collapse as a phase transition in variational autoencoders","authors":"Zhen Li , Fan Zhang , Zheng Zhang , Yu Chen","doi":"10.1016/j.physa.2025.131228","DOIUrl":"10.1016/j.physa.2025.131228","url":null,"abstract":"<div><div>We investigate the phenomenon of posterior collapse in variational autoencoders (VAEs) from the perspective of statistical physics, and reveal that it constitutes a phase transition governed jointly by data structure and model hyper-parameters. By analyzing the stability of the trivial solution associated with posterior collapse, we identify a critical hyper-parameter threshold. In particular, we derive an explicit criterion for the onset of collapse: posterior collapse occurs when the decoder variance exceeds the largest eigenvalue of the data covariance matrix. This critical boundary, separating meaningful latent inference from collapse, is characterized by a discontinuity in the KL-divergence between the approximate posterior and the prior distribution, where the KL-divergence and its derivatives exhibit clear non-analytic behavior. We validate this critical behavior on both synthetic and real-world datasets, confirming the existence of a phase transition. The experimental results align well with our theoretical predictions, demonstrating the robustness of our collapse criterion across various VAE architectures. Our stability-based analysis demonstrate that posterior collapse is not merely an optimization failure, but rather an emerging phase transition arising from the interplay between data structure and variational constraints. This perspective offers new insights into the trainability and representational capacity of deep generative models.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"683 ","pages":"Article 131228"},"PeriodicalIF":3.1,"publicationDate":"2025-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145841030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Zipf-preserving, long-range correlated surrogate for written language and other symbolic sequences 书写语言和其他符号序列的Zipf-preserving，远程相关代理

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2025-12-20 DOI: 10.1016/j.physa.2025.131227

Marcelo A. Montemurro , Mirko Degli Esposti

Symbolic sequences such as written language and genomic DNA display characteristic frequency distributions and long-range correlations extending over many symbols. In language, this takes the form of Zipf’s law for word frequencies together with persistent correlations spanning hundreds or thousands of tokens, while in DNA it is reflected in nucleotide composition and long-memory walks under purine–pyrimidine mappings. Existing surrogate models usually preserve either the frequency distribution or the correlation properties, but not both simultaneously. We introduce a surrogate model that retains both constraints: it preserves the empirical symbol frequencies of the original sequence and reproduces its long-range correlation structure, quantified by the detrended fluctuation analysis (DFA) exponent. Our method generates surrogates of symbolic sequences by mapping fractional Gaussian noise (FGN) onto the empirical histogram through a frequency-preserving assignment. The resulting surrogates match the original in first-order statistics and long-range scaling while randomising short-range dependencies. We validate the model on representative texts in English and Latin, and illustrate its broader applicability with genomic DNA, showing that base composition and DFA scaling are reproduced. This approach provides a principled tool for disentangling structural features of symbolic systems and for testing hypotheses on the origin of scaling laws and memory effects across language, DNA, and other symbolic domains.

符号序列，如书面语言和基因组DNA，显示出特征频率分布和延伸到许多符号的远程相关性。在语言中，这体现为单词频率的齐夫定律，以及跨越数百或数千个标记的持久相关性，而在DNA中，它反映在核苷酸组成和嘌呤-嘧啶映射下的长记忆行走中。现有的代理模型通常要么保留频率分布，要么保留相关性，但不能同时保留两者。我们引入了一个代理模型，保留了这两个约束：它保留了原始序列的经验符号频率，并再现了其远程相关结构，由去趋势波动分析（DFA）指数量化。我们的方法通过保频赋值将分数阶高斯噪声（FGN）映射到经验直方图上，从而生成符号序列的代理。在随机化短期依赖关系的同时，所得到的代理在一阶统计量和长期尺度上与原始匹配。我们在英语和拉丁语的代表性文本上验证了该模型，并说明了其对基因组DNA的更广泛适用性，表明碱基组成和DFA缩放是可复制的。这种方法提供了一个原则性的工具，用于解开符号系统的结构特征，并用于测试关于跨语言、DNA和其他符号领域的缩放定律和记忆效应起源的假设。

{"title":"A Zipf-preserving, long-range correlated surrogate for written language and other symbolic sequences","authors":"Marcelo A. Montemurro , Mirko Degli Esposti","doi":"10.1016/j.physa.2025.131227","DOIUrl":"10.1016/j.physa.2025.131227","url":null,"abstract":"<div><div>Symbolic sequences such as written language and genomic DNA display characteristic frequency distributions and long-range correlations extending over many symbols. In language, this takes the form of Zipf’s law for word frequencies together with persistent correlations spanning hundreds or thousands of tokens, while in DNA it is reflected in nucleotide composition and long-memory walks under purine–pyrimidine mappings. Existing surrogate models usually preserve either the frequency distribution or the correlation properties, but not both simultaneously. We introduce a surrogate model that retains both constraints: it preserves the empirical symbol frequencies of the original sequence and reproduces its long-range correlation structure, quantified by the detrended fluctuation analysis (DFA) exponent. Our method generates surrogates of symbolic sequences by mapping fractional Gaussian noise (FGN) onto the empirical histogram through a frequency-preserving assignment. The resulting surrogates match the original in first-order statistics and long-range scaling while randomising short-range dependencies. We validate the model on representative texts in English and Latin, and illustrate its broader applicability with genomic DNA, showing that base composition and DFA scaling are reproduced. This approach provides a principled tool for disentangling structural features of symbolic systems and for testing hypotheses on the origin of scaling laws and memory effects across language, DNA, and other symbolic domains.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"683 ","pages":"Article 131227"},"PeriodicalIF":3.1,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145841022","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Preserving Microscopic State Integrity for Adaptive Modeling of Complex Spatio-Temporal Systems 保存微观状态完整性的复杂时空系统自适应建模

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2025-12-19 DOI: 10.1016/j.physa.2025.131200

Yue Wang , Yupeng Liu , Chong Wu

Modeling the collective dynamics of complex systems, represented as graph-structured time series, is a central challenge in statistical physics and applied science. Prevailing Spatio-Temporal Graph Neural Networks (STGNNs) are constrained by a serial processing paradigm that creates structural information bottlenecks, leading to the premature abstraction and loss of fine-grained correlations between microscopic system states. This paper posits a hierarchical design principle: the efficacy of adaptive mechanisms, such as attention, is fundamentally contingent upon the integrity of the underlying feature propagation architecture. To validate this, we introduce DMA-EISTGCN, a framework featuring two innovations: (1) an Early Spatio-Temporal Interaction (EI) module, a non-serial design that ensures lossless feature fusion at the model’s front-end, and (2) a Dynamic Multi-scale Attention (DMA) mechanism that adaptively arbitrates between channel, spatial, and regional feature refinement. Experiments on four real-world traffic forecasting benchmarks show the model establishes a new state-of-the-art, reducing prediction error by up to 14.0 % compared to strong SOTA baselines. Critically, an ablation study reveals that applying the adaptive attention mechanism to a conventional serial backbone degrades performance. This provides direct empirical validation for our central thesis. This work establishes a new design paradigm for spatio-temporal models, demonstrating that resolving architectural bottlenecks is a necessary precursor to unlocking the full potential of adaptive learning in complex forecasting tasks.

复杂系统的集体动力学建模，以图结构时间序列表示，是统计物理学和应用科学的核心挑战。当前流行的时空图神经网络（stgnn）受到串行处理范式的限制，这种范式会造成结构信息瓶颈，导致微观系统状态之间的过早抽象和细粒度相关性的丧失。本文假设了一个分层设计原则：自适应机制（如注意力）的有效性从根本上取决于底层特征传播架构的完整性。为了验证这一点，我们引入了DMA- eistgcn框架，该框架具有两个创新：(1)早期时空交互（EI）模块，这是一种确保模型前端无损特征融合的非串行设计；(2)动态多尺度注意（DMA）机制，可自适应地在信道、空间和区域特征细化之间进行仲裁。在四个现实世界交通预测基准上的实验表明，该模型建立了一种新的最先进的技术，与强大的SOTA基线相比，预测误差减少了14.0 %。关键的是，一项消融研究表明，将自适应注意机制应用于传统的串行骨干会降低性能。这为我们的中心论点提供了直接的经验验证。这项工作为时空模型建立了一个新的设计范式，表明解决架构瓶颈是在复杂预测任务中释放自适应学习的全部潜力的必要前提。

{"title":"Preserving Microscopic State Integrity for Adaptive Modeling of Complex Spatio-Temporal Systems","authors":"Yue Wang , Yupeng Liu , Chong Wu","doi":"10.1016/j.physa.2025.131200","DOIUrl":"10.1016/j.physa.2025.131200","url":null,"abstract":"<div><div>Modeling the collective dynamics of complex systems, represented as graph-structured time series, is a central challenge in statistical physics and applied science. Prevailing Spatio-Temporal Graph Neural Networks (STGNNs) are constrained by a serial processing paradigm that creates structural information bottlenecks, leading to the premature abstraction and loss of fine-grained correlations between microscopic system states. This paper posits a hierarchical design principle: the efficacy of adaptive mechanisms, such as attention, is fundamentally contingent upon the integrity of the underlying feature propagation architecture. To validate this, we introduce DMA-EISTGCN, a framework featuring two innovations: (1) an Early Spatio-Temporal Interaction (EI) module, a non-serial design that ensures lossless feature fusion at the model’s front-end, and (2) a Dynamic Multi-scale Attention (DMA) mechanism that adaptively arbitrates between channel, spatial, and regional feature refinement. Experiments on four real-world traffic forecasting benchmarks show the model establishes a new state-of-the-art, reducing prediction error by up to 14.0 % compared to strong SOTA baselines. Critically, an ablation study reveals that applying the adaptive attention mechanism to a conventional serial backbone degrades performance. This provides direct empirical validation for our central thesis. This work establishes a new design paradigm for spatio-temporal models, demonstrating that resolving architectural bottlenecks is a necessary precursor to unlocking the full potential of adaptive learning in complex forecasting tasks.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"684 ","pages":"Article 131200"},"PeriodicalIF":3.1,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145886373","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An analytical insight into the impact of disease-induced mortality on spatiotemporal patterns on directed networks 疾病导致的死亡对定向网络时空格局影响的分析

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2025-12-19 DOI: 10.1016/j.physa.2025.131226

Siqi Duan , Lili Chang , Xing Li , Runzi He , Yihong Li , Xiangfeng Dai , Xiaofeng Luo , Gui-Quan Sun , Zhen Jin

Directed population movement and disease-induced mortality critically influence the spread of infectious diseases. Network reaction diffusion equations offer a powerful tool for characterizing spatiotemporal pattern dynamics. Nevertheless, systematic analytical studies on epidemic patterns in directed networks remain notably scarce. To address this gap, this research aims to develop a theoretical framework for analytically characterizing spatiotemporal patterns in reaction–diffusion epidemic models defined on directed networks, with particular emphasis on disease-induced mortality as a key epidemiological parameter. The primary methodological innovation consists in deriving explicit analytical expressions for the spatiotemporal period and amplitude of emerging patterns through weakly nonlinear analysis conducted near the Turing instability threshold. This analytical approach enables quantitative prediction of pattern characteristics directly from model parameters, overcoming the limitations of purely numerical investigations. Numerical simulations validate the accuracy of these solutions and demonstrate their ability to capture key dynamical features in parameter regimes where mortality significantly influences pattern formation. The results provide a theoretical basis for interpreting periodic transmission patterns of pathogens such as the highly pathogenic avian influenza H5N1 virus in spatially structured populations with directional movement.

人口定向流动和疾病导致的死亡率严重影响传染病的传播。网络反应扩散方程是表征时空格局动力学的有力工具。然而，关于定向网络中流行病模式的系统分析研究仍然非常少。为了解决这一差距，本研究旨在建立一个理论框架，用于分析表征定向网络上定义的反应-扩散流行病模型的时空模式，特别强调疾病引起的死亡率是一个关键的流行病学参数。主要的方法创新在于通过在图灵不稳定阈值附近进行的弱非线性分析，推导出新兴模式的时空周期和幅度的明确解析表达式。这种分析方法能够直接从模型参数中定量预测模式特征，克服了纯粹数值研究的局限性。数值模拟验证了这些解决方案的准确性，并证明了它们在死亡率显著影响模式形成的参数体系中捕捉关键动态特征的能力。研究结果为解释高致病性禽流感H5N1病毒等病原体在具有定向运动的空间结构种群中的周期性传播模式提供了理论基础。

{"title":"An analytical insight into the impact of disease-induced mortality on spatiotemporal patterns on directed networks","authors":"Siqi Duan , Lili Chang , Xing Li , Runzi He , Yihong Li , Xiangfeng Dai , Xiaofeng Luo , Gui-Quan Sun , Zhen Jin","doi":"10.1016/j.physa.2025.131226","DOIUrl":"10.1016/j.physa.2025.131226","url":null,"abstract":"<div><div>Directed population movement and disease-induced mortality critically influence the spread of infectious diseases. Network reaction diffusion equations offer a powerful tool for characterizing spatiotemporal pattern dynamics. Nevertheless, systematic analytical studies on epidemic patterns in directed networks remain notably scarce. To address this gap, this research aims to develop a theoretical framework for analytically characterizing spatiotemporal patterns in reaction–diffusion epidemic models defined on directed networks, with particular emphasis on disease-induced mortality as a key epidemiological parameter. The primary methodological innovation consists in deriving explicit analytical expressions for the spatiotemporal period and amplitude of emerging patterns through weakly nonlinear analysis conducted near the Turing instability threshold. This analytical approach enables quantitative prediction of pattern characteristics directly from model parameters, overcoming the limitations of purely numerical investigations. Numerical simulations validate the accuracy of these solutions and demonstrate their ability to capture key dynamical features in parameter regimes where mortality significantly influences pattern formation. The results provide a theoretical basis for interpreting periodic transmission patterns of pathogens such as the highly pathogenic avian influenza H5N1 virus in spatially structured populations with directional movement.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"683 ","pages":"Article 131226"},"PeriodicalIF":3.1,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145841084","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Competing active and passive rectification mechanisms in a discrete ratchet model of run-and-tumble particles 离散棘轮模型中相互竞争的主动和被动整流机制

IF 3.1 3区物理与天体物理 Q2 PHYSICS, MULTIDISCIPLINARY

Physica A: Statistical Mechanics and its Applications

Pub Date : 2025-12-18 DOI: 10.1016/j.physa.2025.131214

Mesfin Asfaw Taye

We investigate the thermodynamic behavior of a hybrid Brownian motor operating in a discrete three-state ratchet potential, where a particle, modeled as a run-and-tumble active agent, is subject to both thermal asymmetry and internal active propulsion. The model combines features of passive heat engines and active matter systems by allowing the particle to self-propel with velocity

v_{0}

and to switch between internal orientations

σ = \pm 1

at generally asymmetric rates

α_{+}

and

α_{-}

. We derive exact time-dependent and steady-state solutions for the probability distributions, particle current, entropy, and entropy production. Our analysis reveals that, while asymmetric switching breaks time-reversal symmetry and modulates current magnitude, it does not by itself induce current reversal. The direction of transport emerges from the interplay between propulsion, thermal gradients, ratchet asymmetry, and load. We show that the system interpolates continuously between passive and active motor regimes: in the absence of propulsion, the efficiency is bounded by the Carnot limit; with propulsion, it can approach unity in the quasistatic limit. This work provides the first exact time-resolved solution of a discrete-state active ratchet model, offering quantitative insights into nonequilibrium energy conversion in systems that integrate active matter dynamics with passive rectification.

我们研究了在离散三态棘轮势下运行的混合布朗电机的热力学行为，其中粒子被建模为奔跑和翻滚的活性物质，受到热不对称和内部主动推进的影响。该模型结合了被动热机和活性物质系统的特点，允许粒子以速度v0自推进，并以一般不对称的速率α+和α−在内部取向σ=±1之间切换。我们推导出概率分布、粒子电流、熵和熵产生的精确时变和稳态解。我们的分析表明，虽然非对称开关打破了时间反转对称性并调制了电流大小，但它本身并不诱导电流反转。输运方向由推进力、热梯度、棘轮不对称性和载荷之间的相互作用决定。我们证明了系统在被动和主动电机状态之间连续内插：在没有推进的情况下，效率受到卡诺极限的限制；有了推进，它可以在准静态极限下接近统一。这项工作提供了离散状态主动棘轮模型的第一个精确的时间分辨解，为将主动物质动力学与被动整流相结合的系统中的非平衡能量转换提供了定量的见解。

{"title":"Competing active and passive rectification mechanisms in a discrete ratchet model of run-and-tumble particles","authors":"Mesfin Asfaw Taye","doi":"10.1016/j.physa.2025.131214","DOIUrl":"10.1016/j.physa.2025.131214","url":null,"abstract":"<div><div>We investigate the thermodynamic behavior of a hybrid Brownian motor operating in a discrete three-state ratchet potential, where a particle, modeled as a run-and-tumble active agent, is subject to both thermal asymmetry and internal active propulsion. The model combines features of passive heat engines and active matter systems by allowing the particle to self-propel with velocity <span><math><msub><mrow><mi>v</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span> and to switch between internal orientations <span><math><mrow><mi>σ</mi><mo>=</mo><mo>±</mo><mn>1</mn></mrow></math></span> at generally asymmetric rates <span><math><msub><mrow><mi>α</mi></mrow><mrow><mo>+</mo></mrow></msub></math></span> and <span><math><msub><mrow><mi>α</mi></mrow><mrow><mo>−</mo></mrow></msub></math></span>. We derive exact time-dependent and steady-state solutions for the probability distributions, particle current, entropy, and entropy production. Our analysis reveals that, while asymmetric switching breaks time-reversal symmetry and modulates current magnitude, it does not by itself induce current reversal. The direction of transport emerges from the interplay between propulsion, thermal gradients, ratchet asymmetry, and load. We show that the system interpolates continuously between passive and active motor regimes: in the absence of propulsion, the efficiency is bounded by the Carnot limit; with propulsion, it can approach unity in the quasistatic limit. This work provides the first exact time-resolved solution of a discrete-state active ratchet model, offering quantitative insights into nonequilibrium energy conversion in systems that integrate active matter dynamics with passive rectification.</div></div>","PeriodicalId":20152,"journal":{"name":"Physica A: Statistical Mechanics and its Applications","volume":"683 ","pages":"Article 131214"},"PeriodicalIF":3.1,"publicationDate":"2025-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145841024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0