Learning Dissipative Dynamics in Chaotic Systems

被引:0
|
作者
Li, Zongyi [1 ]
Liu-Schiaffini, Miguel [1 ]
Kovachki, Nikola [2 ]
Liu, Burigede [3 ]
Azizzadenesheli, Kamyar [2 ]
Bhattacharya, Kaushik [1 ]
Stuart, Andrew [1 ]
Anandkumar, Anima [1 ]
机构
[1] Caltech, Pasadena, CA 91125 USA
[2] NVIDIA, Santa Clara, CA USA
[3] Univ Cambridge, Cambridge, England
关键词
UNIVERSAL APPROXIMATION;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Chaotic systems are notoriously challenging to predict because of their sensitivity to perturbations and errors due to time stepping. Despite this unpredictable behavior, for many dissipative systems the statistics of the long term trajectories are governed by an invariant measure supported on a set, known as the global attractor; for many problems this set is finite dimensional, even if the state space is infinite dimensional. For Markovian systems, the statistical properties of long-term trajectories are uniquely determined by the solution operator that maps the evolution of the system over arbitrary positive time increments. In this work, we propose a machine learning framework to learn the underlying solution operator for dissipative chaotic systems, showing that the resulting learned operator accurately captures short-time trajectories and long-time statistical behavior. Using this framework, we are able to predict various statistics of the invariant measure for the turbulent Kolmogorov Flow dynamics with Reynolds numbers up to 5000.
引用
收藏
页数:14
相关论文
共 50 条
  • [41] PROJECTION DYNAMICS OF HIGHLY DISSIPATIVE SYSTEMS
    LUBASHEVSKII, IA
    GAFIYCHUK, VV
    PHYSICAL REVIEW E, 1994, 50 (01): : 171 - 181
  • [42] Dissipative particle dynamics for interacting systems
    Pagonabarraga, I
    Frenkel, D
    JOURNAL OF CHEMICAL PHYSICS, 2001, 115 (11): : 5015 - 5026
  • [43] Dissipative dynamics of highly anisotropic systems
    Martinez, Mauricio
    Strickland, Michael
    NUCLEAR PHYSICS A, 2010, 848 (1-2) : 183 - 197
  • [44] Complexity in Hamiltonian-driven dissipative chaotic dynamical systems
    Lai, YC
    Grebogi, C
    PHYSICAL REVIEW E, 1996, 54 (05): : 4667 - 4675
  • [47] Energy-like functions for some dissipative chaotic systems
    Sarasola, C
    D'Anjou, A
    Torrealdea, FJ
    Moujahid, A
    INTERNATIONAL JOURNAL OF BIFURCATION AND CHAOS, 2005, 15 (08): : 2507 - 2521
  • [49] Relatively robust classical structures in dissipative quantum chaotic systems
    Raviola, Lisandro A.
    Carlo, Gabriel G.
    Rivas, Alejandro M. F.
    PHYSICAL REVIEW E, 2010, 81 (04):
  • [50] Dissipative synchronization of nonlinear chaotic systems under information constraints
    Gao, Yanbo
    Zhang, Xiaomei
    Lu, Guoping
    INFORMATION SCIENCES, 2013, 225 : 81 - 97