Discovering interpretable low-dimensional dynamics using maximum entropy
Abstract
Models (i.e., governing equations) are fundamental to science and engineering. Advances in data acquisition now make it possible to extract interpretable, low dimensional descriptions from high dimensional observations. However, existing approaches sacrifice either interpretability for reconstruction accuracy or infer symbolic dynamics without relating latent coordinates to physically meaningful observables. Here we present Edwin (maximum entropy driven compression with interpretable nonlinear model discovery), a unified framework that simultaneously performs dimensionality reduction using the dynamic maximum entropy (DME) principle and discovers sparse symbolic models governing latent dynamics, as well as the coupling between learned features and external metadata. We validate Edwin on diverse simulated systems, including stochastic diffusion, the Ornstein-Uhlenbeck process, self assembling particles, spiking neural populations, and low rank recurrent neural networks, as well as on a noisy experimental time series of aggregating RNA-liposome complexes. Across all systems, Edwin recovers low dimensional symbolic models that are physically interpretable and generalize to unseen conditions. Together, these results establish Edwin as a powerful framework for inferring interpretable, low dimensional dynamics directly from high dimensional data.
Versions
Related
- Discovering compositional, interpretable structure in social behavior
- High-dimensional population codes reveal interpretable and diverse features underlying visual perception
- Minkowski Tensors as a Lightweight and Interpretable Representation for Three-Dimensional Morphology
- FlowMap: Geometry Dynamics Consistent Embedding of RNA Velocity for Interpretable Cellular Trajectories
- Dynamics and geometry of emotion and cognition: an interpretable model of individual human behavior
BibTeX
@misc{chung2026discovering,
title = {{Discovering interpretable low-dimensional dynamics using maximum entropy}},
author = {Michael C. Chung and Tarran Mohan and Purushottam D. Dixit and Juan Guan},
howpublished = {arXiv},
year = {2026},
month = may,
doi = {10.48550/arxiv.2605.16724},
eprint = {2605.16724},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.16724},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Discovering interpretable low-dimensional dynamics using maximum entropy
AU - Michael C. Chung
AU - Tarran Mohan
AU - Purushottam D. Dixit
AU - Juan Guan
PY - 2026
DA - 2026/05/16
PB - arXiv
DO - 10.48550/arxiv.2605.16724
UR - https://arxiv.org/abs/2605.16724
AB - Models (i.e., governing equations) are fundamental to science and engineering. Advances in data acquisition now make it possible to extract interpretable, low dimensional descriptions from high dimensional observations. However, existing approaches sacrifice either interpretability for reconstruction accuracy or infer symbolic dynamics without relating latent coordinates to physically meaningful observables. Here we present Edwin (maximum entropy driven compression with interpretable nonlinear model discovery), a unified framework that simultaneously performs dimensionality reduction using the dynamic maximum entropy (DME) principle and discovers sparse symbolic models governing latent dynamics, as well as the coupling between learned features and external metadata. We validate Edwin on diverse simulated systems, including stochastic diffusion, the Ornstein-Uhlenbeck process, self assembling particles, spiking neural populations, and low rank recurrent neural networks, as well as on a noisy experimental time series of aggregating RNA-liposome complexes. Across all systems, Edwin recovers low dimensional symbolic models that are physically interpretable and generalize to unseen conditions. Together, these results establish Edwin as a powerful framework for inferring interpretable, low dimensional dynamics directly from high dimensional data.
ER -