CKNet: A Convolutional Neural Network Based on Koopman Operator for Modeling Latent Dynamics from Pixels

02/19/2021
by   Yongqian Xiao, et al.
0

For systems with only known pixels, it is difficult to identify its dynamics, especially with a linear operator. In this work, we present a convolutional neural network (CNN) based on the Koopman operator (CKNet) to identify the latent dynamics from raw pixels. CKNet learned an encoder and decoder to play the role of the Koopman eigenfunctions and modes, respectively. The Koopman eigenvalues can be approximated by the eigenvalues of the learned system matrix. We present the deterministic and variational approaches to realize the encoder separately. Because CKNet is trained under the constraints of the Koopman theory, the identified dynamics is linear, controllable and physically-interpretable. Besides, the system matrix and control matrix are trained as trainable tensors. To improve the performance, we propose the auxiliary weight term for multi-step linearity and prediction losses. Experiments select two classic forced dynamical systems with continuous action space, and the results show that identified dynamics with 32-dim can predict validly 120 steps and generate clear images.

READ FULL TEXT
research
11/12/2022

Learning dynamical systems: an example from open quantum system dynamics

Machine learning algorithms designed to learn dynamical systems from dat...
research
10/21/2019

Variational Integrator Networks for Physically Meaningful Embeddings

Learning workable representations of dynamical systems is becoming an in...
research
08/16/2022

Data-driven End-to-end Learning of Pole Placement Control for Nonlinear Dynamics via Koopman Invariant Subspaces

We propose a data-driven method for controlling the frequency and conver...
research
02/03/2023

Koopman Operator Learning: Sharp Spectral Rates and Spurious Eigenvalues

Non-linear dynamical systems can be handily described by the associated ...
research
10/19/2020

Extraction of Discrete Spectra Modes from Video Data Using a Deep Convolutional Koopman Network

Recent deep learning extensions in Koopman theory have enabled compact, ...

Please sign up or login with your details

Forgot password? Click here to reset