<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Wendy_WHY_123的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/Wendy_WHY_123</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; Wendy_WHY_123]]></copyright><item><title><![CDATA[基于信息熵的因果检测方法]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/151788679</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/151788679</guid><author>Wendy_WHY_123</author><pubDate>Thu, 18 Sep 2025 13:39:49 +0800</pubDate><description><![CDATA[摘要 本文提出了一种基于信息熵的因果检测方法——转移熵（Transfer Entropy），用于量化动态系统中子系统间的信息传递。传统互信息方法无法区分实际信息交换与由共同历史或输入信号导致的共享信息，而转移熵通过条件化转移概率克服了这一局限。该方法能够识别驱动与响应元素，检测耦合不对称性（因果方向），适用于非线性系统。作者推导了转移熵的理论框架，讨论了数值实现中的核估计等技术问题，并通过时空系统和生理时间序列验证了其有效性。该方法为研究复杂系统（如生理系统）中的信息流动提供了新工具。]]></description><category></category></item><item><title><![CDATA[Imperfection of the convergent cross-mapping method]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/136706716</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/136706716</guid><author>Wendy_WHY_123</author><pubDate>Thu, 14 Mar 2024 18:14:44 +0800</pubDate><description><![CDATA[2012 年，用于从时间序列中发现系统变量之间因果关系的收敛交叉映射法出版。这种方法被广泛应用于各种性质的系统研究–从评估宇宙辐射对气候的影响，到研究大脑活动。使用这种方法的意义和前景促使我们掌握这种方法，并将其用于确定 "生物圈-气候 "系统变量之间的因果关系。得到的结果似乎很可疑，迫使我们仔细检查 CCM 结论与各种模型系统之间的适当性。本文举例说明了该方法对因果关系的存在做出错误估计的模型情况，并对出现错误估计的原因做出了可能的解释。]]></description><category></category></item><item><title><![CDATA[Distance preservation in state-space methods for detecting causal interactions in dynamical systems]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/136391297</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/136391297</guid><author>Wendy_WHY_123</author><pubDate>Wed, 06 Mar 2024 16:19:20 +0800</pubDate><description><![CDATA[我们分析了流行的 "状态空间 "类算法，用于检测耦合动力系统中的偶然相互作用。这些算法通常以塔肯斯嵌入定理为依据，该定理提供了涉及吸引子及其延迟嵌入的关系是连续的条件。然而，在实践中，状态空间方法通常并不直接测试连续性，而是测试这些关系如何保持点间距离这一更强的属性。本文从距离保持的角度出发，对状态空间算法进行了明确的理论和实证探索。首先，我们推导出适用于简单耦合系统的基本理论保证，提供了某些映射的距离保持揭示潜在因果结构的条件。其次，我们通过经验证明，典型的耦合系统并不满足距离保持假设。]]></description><category></category></item><item><title><![CDATA[灵长类动物手臂的计算模型：从手部位置到关节角度、关节扭矩和肌肉力]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/126580947</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/126580947</guid><author>Wendy_WHY_123</author><pubDate>Sat, 03 Sep 2022 10:51:33 +0800</pubDate><description><![CDATA[非人类灵长类动物的三维触及是研究大脑运动控制区域中存在的表征的重要行为范式。迄今为止，大多数研究都将神经活动与许多手臂运动参数中的一些参数相关联，包括：整体手部位置或速度、关节角度、关节角速度、关节扭矩或肌肉激活。到目前为止，还没有一项研究能够以有意义的方式整合所有这些参数，从而分离这些通常高度相关的变量。]]></description><category></category></item><item><title><![CDATA[A Crash Course in Good and Bad Controls]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/126666380</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/126666380</guid><author>Wendy_WHY_123</author><pubDate>Sat, 03 Sep 2022 10:16:54 +0800</pubDate><description><![CDATA[许多统计学和计量经济学的学生对传统文献中处理“不良控制”问题的方式表示失望。当将变量添加到回归方程时，会在回归系数与系数打算表示的效果之间产生意外差异时，就会出现问题。避免这种差异对数据密集型科学中的所有分析师提出了挑战。本说明通过一系列说明性示例描述了用于理解、可视化和解决问题的图形工具。通过让教师和从业者可以使用这个“速成课程”，我们希望将这些工具用于更广泛的关注回归模型因果解释的科学家社区。]]></description><category></category></item><item><title><![CDATA[Learning on knowledge graph dynamics provides an early warning of impactful research]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/119509702</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/119509702</guid><author>Wendy_WHY_123</author><pubDate>Fri, 13 Aug 2021 19:10:56 +0800</pubDate><description><![CDATA[@[TOC](Learning on knowledge graph dynamics provides an early warning of impactful research(学习知识图谱动态为有影响力的研究提供早期预警))
Abstract（摘要）
科学生态系统中评价一个研究的好坏仅依赖于基于引文的指标，这个指标只能提供不完善、不一致且易于操作的研究质量指标。在这里，我们提出了DELPHI（Dynamic Early-warning by Learning to Predict High Impa]]></description><category></category></item><item><title><![CDATA[Unveiling causal interactions in complex systems（揭示复杂系统中的因果交互作用）]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/114883056</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/114883056</guid><author>Wendy_WHY_123</author><pubDate>Tue, 16 Mar 2021 21:08:22 +0800</pubDate><description><![CDATA[Unveiling causal interactions in complex systems摘要引言结果监督生态系统的相互依赖性通过大脑活动诊断疾病监控衍生品的系统性风险追踪入侵动态和天气对沙漠生态系统的影响揭示了酗酒者大脑网络的独特特征在CDS市场中发现持续的因果关系和有影响力的资产讨论方法符号信息因果关系评估的框架吸引子重构最近的邻居和他们的未来预测受影响变量的预测模式h提前一步驱动变量的模式受影响变量的真实模式(回测过程)每一时刻t步的影响性质和强度因果关系的总体性质和强度因果网络分析力量中心链接]]></description><category></category></item><item><title><![CDATA[Detecting Causality in Complex Ecosystems（检测复杂生态系统中的因果关系附件）]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/114802158</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/114802158</guid><author>Wendy_WHY_123</author><pubDate>Mon, 15 Mar 2021 20:47:37 +0800</pubDate><description><![CDATA[Detecting Causality in Complex Ecosystems摘要摘要摘要摘要摘要
作者：George Sugihara, Robert May, Hao Ye,1Chih-hao Hsieh,  Ethan Deyle, Michael Fogarty, Stephan Munch
翻译：Wendy
摘要
  
摘要
  
摘要
  
摘要
  
摘要
  

...]]></description><category></category></item><item><title><![CDATA[Detecting Causality in Complex Ecosystems（检测复杂生态系统中的因果关系）]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/114789368</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/114789368</guid><author>Wendy_WHY_123</author><pubDate>Sun, 14 Mar 2021 20:23:28 +0800</pubDate><description><![CDATA[Detecting Causality in Complex Ecosystems摘要Ⅰ 引言动态因果关系和CCM识别因果关系的框架（i）通过功能耦合进行双向因果关系识别(ii)单向因果关系传递性复杂的模型例子:非耦合变量的外部强迫现实世界的例子:用生态数据进行演示关于不可分离性的最后评论结论
作者：George Sugihara, Robert May, Hao Ye,1Chih-hao Hsieh,  Ethan Deyle, Michael Fogarty, Stephan Munch
翻译：Wend]]></description><category></category></item><item><title><![CDATA[A Survey on Neural Network Interpretability (神经网络的可解释性研究综述)]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/113457295</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/113457295</guid><author>Wendy_WHY_123</author><pubDate>Sun, 14 Mar 2021 15:00:41 +0800</pubDate><description><![CDATA[A Survey on Neural Network Interpretability摘要Ⅰ 引言Ⅱ 引言Ⅲ 引言Ⅳ 引言Ⅴ 引言
作者：Yu Zhang, Peter Tino, Ale s Leonardis, and Ke Tang
翻译：Wendy
摘要
  随着神经网络的巨大成功，人们对其黑匣子的性质也越来越关注。神经网络的可解释性影响着人们对深度学习系统的信赖。它也与许多道德问题有关，例如算法歧视。此外，可解释性是深度神经网络成为其他领域（例如药物研发和基因组学）中强大工具的重要属性。在这篇综述]]></description><category></category></item><item><title><![CDATA[Chaos as an Intermittently Forced Linear System（混沌作为一种间歇强迫线性系统）]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/108702949</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/108702949</guid><author>Wendy_WHY_123</author><pubDate>Wed, 23 Sep 2020 10:32:11 +0800</pubDate><description><![CDATA[Chaos as an Intermittently Forced Linear System摘要1 Introduction（引言）2 Background（背景）2.1 Koopman operator theory（Koopman算子理论）摘要
作者：Steven L. Bruntonn, Bingni W. Brunton, Joshua L. Proctor, Eurika Kaiser, J. Nathan Kutz
翻译：Wendy
摘要
  理解混沌系统中有序和无序的相互作用是现代定量科学的]]></description><category></category></item><item><title><![CDATA[使用几何构造用于确定重构相空间的嵌入大小]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/106443052</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/106443052</guid><author>Wendy_WHY_123</author><pubDate>Mon, 01 Jun 2020 22:33:23 +0800</pubDate><description><![CDATA[Determining embedding dimension for phase-space reconstruction using a geometrical construction摘要Ⅰ 引言（Introduction）Ⅱ 识别假邻居节点的方法（THE METHOD OF FALSE NEIGHBORS）种子集识别对丢失或错误数据的敏感性
作者：Matthew B.Kennel，Reggie Brown ， Henry D. I.Abarbanel
翻译：Wendy
摘要
  我们通过观察在d→d]]></description><category></category></item><item><title><![CDATA[附件-代谢环境的大规模重建和系统发育分析（Large-scale reconstruction and phylogenetic analysis of metabolic environments）]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/105972968</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/105972968</guid><author>Wendy_WHY_123</author><pubDate>Fri, 08 May 2020 21:42:51 +0800</pubDate><description><![CDATA[Large-scale reconstruction and phylogenetic analysis of metabolic environments种子集识别对丢失或错误数据的敏感性紧密连接的组件统计信息紧密连接的组件统计信息
作者：Elhanan Borenstein，Martin Kupiec， Marcus W. Feldman， and Eytan Ruppin
翻译：Wendy
...]]></description><category></category></item><item><title><![CDATA[代谢环境的大规模重建和系统发育分析（Large-scale reconstruction and phylogenetic analysis of metabolic environments）]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/105933259</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/105933259</guid><author>Wendy_WHY_123</author><pubDate>Thu, 07 May 2020 11:21:41 +0800</pubDate><description><![CDATA[Large-scale reconstruction and phylogenetic analysis of metabolic environments摘要Ⅰ.引言Ⅱ.结果（Results）Ⅰ.引言
作者：Elhanan Borenstein，Martin Kupiec， Marcus W. Feldman， and Eytan Ruppin
翻译：Wendy
摘要
  代谢网络的拓扑结构不仅...]]></description><category></category></item><item><title><![CDATA[网络中的模块化和社区结构(Modularity and community structure in networks)]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/105820269</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/105820269</guid><author>Wendy_WHY_123</author><pubDate>Wed, 29 Apr 2020 17:00:30 +0800</pubDate><description><![CDATA[Machine learning and the physical sciences摘要Ⅰ.引言Ⅵ.化学和材料科学（  CHEMISTRY AND MATERIALS）A.基于原子环境的能量和力（ Energies and forces based on atomic environments）
作者：M. E. J. Newman
翻译：Wendy
摘要
  人们发现，许多与科学有关的网络，包括...]]></description><category></category></item><item><title><![CDATA[机器学习与物理科学（五）（Machine learning and the physical sciences）]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/104826538</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/104826538</guid><author>Wendy_WHY_123</author><pubDate>Mon, 16 Mar 2020 11:41:25 +0800</pubDate><description><![CDATA[Machine learning and the physical sciences摘要Ⅲ.粒子物理与宇宙学（ PARTICLE PHYSICS AND COSMOLOGY）A.发展历史（Historical note）1.有助于理解基本的无监督方法 ( Contributions to understanding basic unsupervised methods)Ⅱ.统计物理（STATIST...]]></description><category></category></item><item><title><![CDATA[机器学习与物理科学（四）（Machine learning and the physical sciences）]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/104826525</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/104826525</guid><author>Wendy_WHY_123</author><pubDate>Sat, 14 Mar 2020 11:21:31 +0800</pubDate><description><![CDATA[Machine learning and the physical sciences摘要Ⅲ.粒子物理与宇宙学（ PARTICLE PHYSICS AND COSMOLOGY）A.发展历史（Historical note）1.有助于理解基本的无监督方法 ( Contributions to understanding basic unsupervised methods)Ⅱ.统计物理（STATIST...]]></description><category></category></item><item><title><![CDATA[机器学习与物理科学（三）（Machine learning and the physical sciences）]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/104825788</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/104825788</guid><author>Wendy_WHY_123</author><pubDate>Fri, 13 Mar 2020 12:00:05 +0800</pubDate><description><![CDATA[Machine learning and the physical sciences摘要Ⅱ.统计物理（STATISTICAL PHYSICS）A.发展历史（Historical note）1.有助于理解基本的无监督方法 ( Contributions to understanding basic unsupervised methods)Ⅱ.统计物理（STATISTICAL PHYSICS）A.发...]]></description><category></category></item><item><title><![CDATA[机器学习与物理科学（二）（Machine learning and the physical sciences）]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/104805342</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/104805342</guid><author>Wendy_WHY_123</author><pubDate>Thu, 12 Mar 2020 20:01:42 +0800</pubDate><description><![CDATA[Machine learning and the physical sciences摘要Ⅱ.统计物理（STATISTICAL PHYSICS）A.发展历史（Historical note）B.深度学习中的理论难题（Theoretical puzzles in deep learning）B.无监督学习中的统计物理（ Statistical physics of unsupervised learn...]]></description><category></category></item><item><title><![CDATA[机器学习与物理科学（一）（Machine learning and the physical sciences）]]></title><link>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/104793247</link><guid>https://blog.youkuaiyun.com/Wendy_WHY_123/article/details/104793247</guid><author>Wendy_WHY_123</author><pubDate>Wed, 11 Mar 2020 20:52:18 +0800</pubDate><description><![CDATA[Machine learning and the physical sciences摘要Ⅰ.引言（INTRODUCTION）A. 机器学习相关概念（ Concepts in machine learning）1.监督学习和神经网络（ Supervised learning and neural networks）2.无监督学习和生成建模（Unsupervised learning and gene...]]></description><category></category></item></channel></rss>