Phase2vec: Dynamical systems embedding with a physics-informed convolutional network

• URL: http://arxiv.org/abs/2212.03857v1
• Date: Wed, 7 Dec 2022 18:54:52 GMT
• Title: Phase2vec: Dynamical systems embedding with a physics-informed convolutional network
• Authors: Matthew Ricci, Noa Moriel, Zoe Piran, Mor Nitzan
• Abstract summary: We propose an embedding method that learns high-quality, physically-meaningful representations of 2D dynamical systems without supervision. Our embeddings encode important physical properties of the underlying data, including the stability of fixed points, conservation of energy, and the incompressibility of flows.
• Score: 1.6058099298620423
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Dynamical systems are found in innumerable forms across the physical and biological sciences, yet all these systems fall naturally into universal equivalence classes: conservative or dissipative, stable or unstable, compressible or incompressible. Predicting these classes from data remains an essential open challenge in computational physics at which existing time-series classification methods struggle. Here, we propose, \texttt{phase2vec}, an embedding method that learns high-quality, physically-meaningful representations of 2D dynamical systems without supervision. Our embeddings are produced by a convolutional backbone that extracts geometric features from flow data and minimizes a physically-informed vector field reconstruction loss. In an auxiliary training period, embeddings are optimized so that they robustly encode the equations of unseen data over and above the performance of a per-equation fitting method. The trained architecture can not only predict the equations of unseen data, but also, crucially, learns embeddings that respect the underlying semantics of the embedded physical systems. We validate the quality of learned embeddings investigating the extent to which physical categories of input data can be decoded from embeddings compared to standard blackbox classifiers and state-of-the-art time series classification techniques. We find that our embeddings encode important physical properties of the underlying data, including the stability of fixed points, conservation of energy, and the incompressibility of flows, with greater fidelity than competing methods. We finally apply our embeddings to the analysis of meteorological data, showing we can detect climatically meaningful features. Collectively, our results demonstrate the viability of embedding approaches for the discovery of dynamical features in physical systems.

Related papers

• Learning Physics From Video: Unsupervised Physical Parameter Estimation for Continuous Dynamical Systems [49.11170948406405]
  State-of-the-art in automatic parameter estimation from video is addressed by training supervised deep networks on large datasets. We propose a method to estimate the physical parameters of any known, continuous governing equation from single videos.
  arXiv  Detail & Related papers  (2024-10-02T09:44:54Z)
• Data-Driven Computing Methods for Nonlinear Physics Systems with Geometric Constraints [0.7252027234425334]
  This paper introduces a novel, data-driven framework that synergizes physics-based priors with advanced machine learning techniques. Our framework showcases four algorithms, each embedding a specific physics-based prior tailored to a particular class of nonlinear systems. The integration of these priors also enhances the expressive power of neural networks, enabling them to capture complex patterns typical in physical phenomena.
  arXiv  Detail & Related papers  (2024-06-20T23:10:41Z)
• $Φ$-DVAE: Physics-Informed Dynamical Variational Autoencoders for Unstructured Data Assimilation [3.2873782624127843]
  We develop a physics-informed dynamical variational autoencoder ($Phi$-DVAE) to embed diverse data streams into time-evolving physical systems. Our approach combines a standard, possibly nonlinear, filter for the latent state-space model and a VAE, to assimilate the unstructured data into the latent dynamical system. A variational Bayesian framework is used for the joint estimation of the encoding, latent states, and unknown system parameters.
  arXiv  Detail & Related papers  (2022-09-30T17:34:48Z)
• Capturing Actionable Dynamics with Structured Latent Ordinary Differential Equations [68.62843292346813]
  We propose a structured latent ODE model that captures system input variations within its latent representation. Building on a static variable specification, our model learns factors of variation for each input to the system, thus separating the effects of the system inputs in the latent space.
  arXiv  Detail & Related papers  (2022-02-25T20:00:56Z)
• Leveraging the structure of dynamical systems for data-driven modeling [111.45324708884813]
  We consider the impact of the training set and its structure on the quality of the long-term prediction. We show how an informed design of the training set, based on invariants of the system and the structure of the underlying attractor, significantly improves the resulting models.
  arXiv  Detail & Related papers  (2021-12-15T20:09:20Z)
• Robust physics discovery via supervised and unsupervised pattern recognition using the Euler characteristic [5.584060970507507]
  Machine learning approaches have been widely used for discovering the underlying physics of dynamical systems from measured data. Existing approaches, still lack robustness, especially the measured data contain a large level of noise. In this study, we use an efficient topological descriptor for complex data, ie.temporal characteristics (ECs) as features to characterize the observed data.
  arXiv  Detail & Related papers  (2021-10-15T18:37:42Z)
• An Extensible Benchmark Suite for Learning to Simulate Physical Systems [60.249111272844374]
  We introduce a set of benchmark problems to take a step towards unified benchmarks and evaluation protocols. We propose four representative physical systems, as well as a collection of both widely used classical time-based and representative data-driven methods.
  arXiv  Detail & Related papers  (2021-08-09T17:39:09Z)
• Using Data Assimilation to Train a Hybrid Forecast System that Combines Machine-Learning and Knowledge-Based Components [52.77024349608834]
  We consider the problem of data-assisted forecasting of chaotic dynamical systems when the available data is noisy partial measurements. We show that by using partial measurements of the state of the dynamical system, we can train a machine learning model to improve predictions made by an imperfect knowledge-based model.
  arXiv  Detail & Related papers  (2021-02-15T19:56:48Z)
• Data augmentation and feature selection for automatic model recommendation in computational physics [0.0]
  This article introduces two algorithms to address the lack of training data, their high dimensionality, and the non-applicability of common data augmentation techniques to physics data. When combined with a stacking ensemble made of six multilayer perceptrons and a ridge logistic regression, they enable reaching an accuracy of 90% on our classification problem for nonlinear structural mechanics.
  arXiv  Detail & Related papers  (2021-01-12T15:09:11Z)
• Active Learning for Nonlinear System Identification with Guarantees [102.43355665393067]
  We study a class of nonlinear dynamical systems whose state transitions depend linearly on a known feature embedding of state-action pairs. We propose an active learning approach that achieves this by repeating three steps: trajectory planning, trajectory tracking, and re-estimation of the system from all available data. We show that our method estimates nonlinear dynamical systems at a parametric rate, similar to the statistical rate of standard linear regression.
  arXiv  Detail & Related papers  (2020-06-18T04:54:11Z)
• Physics-informed learning of governing equations from scarce data [14.95055620484844]
  This work introduces a physics-informed deep learning framework to discover governing partial differential equations (PDEs) from scarce and noisy representation data. The efficacy and robustness of this method are demonstrated, both numerically and experimentally, on discovering a variety of PDE systems. The resulting computational framework shows the potential for closed-form model discovery in practical applications.
  arXiv  Detail & Related papers  (2020-05-05T22:13:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.