.png){kind=icon}
.png){kind=icon}
Return
A novel kernelized angle metric for state similarity in nonlinear dynamic systems
.png){kind=icon}
PRE
AI
Z
屈
S
DOI：10.1016/j.chaos.2025.117188.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A novel Kernelized Angle Metric (KAM) combines distance and direction for state similarity. • KAM consistently outperforms Euclidean distance, demonstrating robustness against high observational noise. • The effectiveness of KAM is algorithm-dependent: it dramatically enhances simpler models like Simplex. • KAM provides an efficient, powerful alternative to manifold distance.
Journal
C
IF:
5.6
Papers:
1.3K
Citations:
3.8W
Organization
Cited Papers