符号动力学
混乱的
相空间
统计物理学
系列(地层学)
配分函数(量子场论)
分拆(数论)
计算机科学
符号数据分析
时间序列
算法
数学
物理
理论计算机科学
纯数学
人工智能
量子力学
机器学习
古生物学
组合数学
生物
作者
Erik M. Bollt,Theodore Stanford,Ying‐Cheng Lai,Karol Życzkowski
标识
DOI:10.1103/physrevlett.85.3524
摘要
A practical and popular technique to extract the symbolic dynamics from experimentally measured chaotic time series is the threshold-crossing method, by which an arbitrary partition is utilized for determining the symbols. We address to what extent the symbolic dynamics so obtained can faithfully represent the phase-space dynamics. Our principal result is that such practice leads to a severe misrepresentation of the dynamical system. The measured topological entropy is a Devil's staircase-like, but surprisingly nonmonotone, function of a parameter characterizing the amount of misplacement of the partition.
科研通智能强力驱动
Strongly Powered by AbleSci AI