Information bottleneck for learning the phase space of dynamics from high-dimensional experimental data

• URL: http://arxiv.org/abs/2604.24662v1
• Date: Mon, 27 Apr 2026 16:24:45 GMT
• Title: Information bottleneck for learning the phase space of dynamics from high-dimensional experimental data
• Abstract summary: We introduce DySIB as a method to learn low-dimensional representations of time-series data.<n>We apply DySIB to an experimental video dataset of a pendulum, where the underlying state space is known.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Identifying the dynamical state variables of a system from high-dimensional observations is a central problem across physical sciences. The challenge is that the state variables are not directly observable and must be inferred from raw high-dimensional data without supervision. Here we introduce DySIB (Dynamical Symmetric Information Bottleneck) as a method to learn low-dimensional representations of time-series data by maximizing predictive mutual information between past and future observation windows while penalizing representation complexity. This objective operates entirely in latent space and avoids reconstruction of the observations. We apply DySIB to an experimental video dataset of a physical pendulum, where the underlying state space is known. The method, with hyperparameters of the learning architecture set self-consistently by the data, recovers a two-dimensional representation that matches the dimensionality, topology, and geometry of the pendulum phase space, with the learned coordinates aligning smoothly with the canonical angle and angular velocity. These results demonstrate, on a well-characterized experimental system, that predictive information in latent space can be used to recover interpretable dynamical coordinates directly from high-dimensional data.

Related papers

• DyG$^2$T: Modeling Object Dynamics with 3D Gaussian Temporal-Spatial Particle Graph Transformer [86.4618122245946]
  DyG$2$T is a dynamics modeling framework that infers object motion trajectories by spatially completing and temporally discriminating Key Point representations.<n>Experiments on both synthetic and real-world datasets demonstrate that DyG$2$T achieves accurate dynamics modeling and reasoning.
  arXiv  Detail & Related papers  (2026-08-19T03:44:28Z)
• VAE with Hyperspherical Coordinates: Improving Anomaly Detection from Hypervolume-Compressed Latent Space [56.362776482614976]
  Variational autoencoders (VAE) encode data into lower-dimensional latent vectors before decoding those vectors back to data.<n>We propose to formulate the latent variables of a VAE using hyperspherical coordinates, which allows compressing the latent vectors towards a given direction on the hypersphere.<n>We show that this improves both the fully unsupervised and OOD anomaly detection ability of the VAE, achieving the best performance on the datasets we considered.
  arXiv  Detail & Related papers  (2026-01-25T03:10:24Z)
• Machine Learning for Scientific Visualization: Ensemble Data Analysis [0.0]
  This dissertation explores deep learning techniques to improve scientific visualization.<n>We introduce autoencoder-supervised dimensionality reduction for scientific ensembles.<n>Next, we present FLINT, a deep learning model for expressive high-quality flow estimation.<n>Finally, we introduce HyperFLINT, a hypernetwork-based approach to estimate flow fields and interpolate data.
  arXiv  Detail & Related papers  (2025-11-28T15:45:54Z)
• Geometric Operator Learning with Optimal Transport [77.16909146519227]
  We propose integrating optimal transport (OT) into operator learning for partial differential equations (PDEs) on complex geometries.<n>For 3D simulations focused on surfaces, our OT-based neural operator embeds the surface geometry into a 2D parameterized latent space.<n> Experiments with Reynolds-averaged Navier-Stokes equations (RANS) on the ShapeNet-Car and DrivAerNet-Car datasets show that our method achieves better accuracy and also reduces computational expenses.
  arXiv  Detail & Related papers  (2025-07-26T21:28:25Z)
• Follow the Energy, Find the Path: Riemannian Metrics from Energy-Based Models [63.331590876872944]
  We propose a method for deriving Riemannian metrics directly from pretrained Energy-Based Models.<n>These metrics define spatially varying distances, enabling the computation of geodesics.<n>We show that EBM-derived metrics consistently outperform established baselines.
  arXiv  Detail & Related papers  (2025-05-23T12:18:08Z)
• Geometry-aware Active Learning of Spatiotemporal Dynamic Systems [4.251030047034566]
  This paper proposes a geometry-aware active learning framework for modeling dynamic systems.<n>We develop an adaptive active learning strategy to strategically identify spatial locations for data collection and further maximize the prediction accuracy.
  arXiv  Detail & Related papers  (2025-04-26T19:56:38Z)
• Ultralow-dimensionality reduction for identifying critical transitions by spatial-temporal PCA [8.474631244771928]
  stPCA represents dynamics of a high-dimensional time-series by only a single latent variable without distortion.<n>The dynamics of this single variable is analytically solved and theoretically preserves the temporal property of original time-series.<n>Its applications to real-world datasets demonstrated the effectiveness of stPCA.
  arXiv  Detail & Related papers  (2025-01-22T02:09:52Z)
• Conservation-informed Graph Learning for Spatiotemporal Dynamics Prediction [84.26340606752763]
  In this paper, we introduce the conservation-informed GNN (CiGNN), an end-to-end explainable learning framework.<n>The network is designed to conform to the general symmetry conservation law via symmetry where conservative and non-conservative information passes over a multiscale space by a latent temporal marching strategy.<n>Results demonstrate that CiGNN exhibits remarkable baseline accuracy and generalizability, and is readily applicable to learning for prediction of varioustemporal dynamics.
  arXiv  Detail & Related papers  (2024-12-30T13:55:59Z)
• Noisy Data Visualization using Functional Data Analysis [14.255424476694946]
  We propose a new data visualization method called Functional Information Geometry (FIG) for dynamical processes. We experimentally demonstrate that the resulting method outperforms a variant of EIG designed for visualization. We then use our method to visualize EEG brain measurements of sleep activity.
  arXiv  Detail & Related papers  (2024-06-05T15:53:25Z)
• Deep Kernel Learning of Dynamical Models from High-Dimensional Noisy Data [1.3750624267664155]
  The framework is composed of an encoder that compresses high-dimensional measurements into low-dimensional state variables. The training of the proposed model is carried out in an unsupervised manner, not relying on labeled data. Results show that the method can effectively denoise measurements, learn compact state representations and latent dynamical models.
  arXiv  Detail & Related papers  (2022-08-27T09:47:44Z)
• Towards Scale Consistent Monocular Visual Odometry by Learning from the Virtual World [83.36195426897768]
  We propose VRVO, a novel framework for retrieving the absolute scale from virtual data. We first train a scale-aware disparity network using both monocular real images and stereo virtual data. The resulting scale-consistent disparities are then integrated with a direct VO system.
  arXiv  Detail & Related papers  (2022-03-11T01:51:54Z)
• Visualizing the Finer Cluster Structure of Large-Scale and High-Dimensional Data [7.400745342582259]
  We propose using a generalized sigmoid function to model the distance similarity in both high- and low-dimensional spaces. Using both simulated and real-world data sets, we show that our proposed method can generate visualization results comparable to those of uniform manifold approximation and projection.
  arXiv  Detail & Related papers  (2020-07-17T01:36:45Z)

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.