Eigenvalue initialisation and regularisation for Koopman autoencoders

TL;DR

This paper introduces eigenvalue-based initialisation and regularisation schemes for Koopman autoencoders, improving training efficiency and prediction accuracy in physics-related problems.

Contribution

It proposes eigeninit and eigenloss schemes tailored for Koopman autoencoders, enhancing their performance in physics applications.

Findings

Eigenloss and eigeninit improve convergence rate by up to 5 times.

They reduce long-term prediction error by up to 3 times.

Applicable to synthetic and real-world physics datasets.

Abstract

Regularising the parameter matrices of neural networks is ubiquitous in training deep models. Typical regularisation approaches suggest initialising weights using small random values, and to penalise weights to promote sparsity. However, these widely used techniques may be less effective in certain scenarios. Here, we study the Koopman autoencoder model which includes an encoder, a Koopman operator layer, and a decoder. These models have been designed and dedicated to tackle physics-related problems with interpretable dynamics and an ability to incorporate physics-related constraints. However, the majority of existing work employs standard regularisation practices. In our work, we take a step toward augmenting Koopman autoencoders with initialisation and penalty schemes tailored for physics-related settings. Specifically, we propose the "eigeninit" initialisation scheme that samples initial Koopman operators from specific eigenvalue distributions. In addition, we suggest the "eigenloss" penalty scheme that penalises the eigenvalues of the Koopman operator during training. We demonstrate the utility of these schemes on two synthetic data sets: a driven pendulum and flow past a cylinder; and two real-world problems: ocean surface temperatures and cyclone wind fields. We find on these datasets that eigenloss and eigeninit improves the convergence rate by up to a factor of 5, and that they reduce the cumulative long-term prediction error by up to a factor of 3. Such a finding points to the utility of incorporating similar schemes as an inductive bias in other physics-related deep learning approaches.