DeepAI AI Chat
Log In Sign Up

Markov Neural Operators for Learning Chaotic Systems

06/13/2021
by   Zongyi Li, et al.
16

Chaotic systems are notoriously challenging to predict because of their instability. Small errors accumulate in the simulation of each time step, resulting in completely different trajectories. However, the trajectories of many prominent chaotic systems live in a low-dimensional subspace (attractor). If the system is Markovian, the attractor is uniquely determined by the Markov operator that maps the evolution of infinitesimal time steps. This makes it possible to predict the behavior of the chaotic system by learning the Markov operator even if we cannot predict the exact trajectory. Recently, a new framework for learning resolution-invariant solution operators for PDEs was proposed, known as neural operators. In this work, we train a Markov neural operator (MNO) with only the local one-step evolution information. We then compose the learned operator to obtain the global attractor and invariant measure. Such a Markov neural operator forms a discrete semigroup and we empirically observe that does not collapse or blow up. Experiments show neural operators are more accurate and stable compared to previous methods on chaotic systems such as the Kuramoto-Sivashinsky and Navier-Stokes equations.

READ FULL TEXT

page 6

page 10

page 11

page 13

02/12/2023

Physics informed WNO

Deep neural operators are recognized as an effective tool for learning s...
08/02/2022

Approximate Bayesian Neural Operators: Uncertainty Quantification for Parametric PDEs

Neural operators are a type of deep architecture that learns to solve (i...
07/15/2021

On universal approximation and error bounds for Fourier Neural Operators

Fourier neural operators (FNOs) have recently been proposed as an effect...
11/17/2020

Splitting Schemes for Some Second-Order Evolution Equations

We consider the Cauchy problem for a second-order evolution equation, in...
09/28/2011

Admissible and Restrained Revision

As partial justification of their framework for iterated belief revision...
10/03/2022

Nonlinear Reconstruction for Operator Learning of PDEs with Discontinuities

A large class of hyperbolic and advection-dominated PDEs can have soluti...