Identifiable learning of dissipative dynamics
Not on the map. Why?
Abstract (arXiv)
Complex dissipative systems appear across science and engineering, from polymers and active matter to learning algorithms. These systems operate far from equilibrium, where energy dissipation and time irreversibility govern their behavior but are difficult to quantify from data. Here, we introduce a universal and identifiable neural framework that learns dissipative stochastic dynamics directly from trajectories while ensuring interpretability, expressiveness, and uniqueness. Our method identifies a unique energy landscape, separates reversible from irreversible motion, and allows direct computation of the entropy production, providing a principled measure of irreversibility and deviations from equilibrium. Applications to polymer stretching in elongational flow and to stochastic gradient Langevin dynamics reveal new insights, including super-linear scaling of barrier heights and sub-linear scaling of entropy production rates with the strain rate, and the suppression of irreversibility with increasing batch size. Our methodology thus establishes a general, data-driven framework for discovering and interpreting non-equilibrium dynamics.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 23 of its 94 references that are in the map. First posted on arXiv 28 Oct 2025.
Builds on
23 of 94 references.
References · 23
- 2018Neural Ordinary Differential Equationscited by 3,779
- 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,908
- 2019DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operatorscited by 2,018
- 2019Hamiltonian Neural Networkscited by 712
- 2018Stochastic Modified Equations and Dynamics of Stochastic Gradient Algorithms I: Mathematical Foundationscited by 114
- 2017Stochastic gradient descent performs variational inference, converges to limit cycles for deep networkscited by 109
- 2020AI Poincar'e: Machine Learning Conservation Laws from Trajectoriescited by 77
- 2021Port-Hamiltonian Neural Networks for Learning Explicit Time-Dependent Dynamical Systemscited by 67
- 2021Machine learning structure preserving brackets for forecasting irreversible processescited by 52
- 2021GFINNs: GENERIC Formalism Informed Neural Networks for Deterministic and Stochastic Dynamical Systemscited by 49
- 2021Learning effective stochastic differential equations from microscopic simulations: linking stochastic numerics to deep learningcited by 39
- 2020OnsagerNet: Learning Stable and Interpretable Dynamics using a Generalized Onsager Principlecited by 37
11 more
- 2024tLaSDI: Thermodynamics-informed latent space dynamics identificationcited by 23
- 2023Constructing Custom Thermodynamics Using Deep Learningcited by 17
- 2023Deep learning probability flows and entropy production rates in active mattercited by 15
- 2022Neural parameter calibration for large-scale multi-agent modelscited by 14
- 2024DynGMA: a robust approach for learning stochastic differential equations from datacited by 8
- 2025DUE: A Deep Learning Framework and Library for Modeling Unknown Equationscited by 8
- 2023Inferring networks from time series: a neural approachcited by 7
- 2021Computing the Invariant Distribution of Randomly Perturbed Dynamical Systems Using Deep Learningcited by 4
- 2025Nonequilibrium physics of brain dynamicscited by 4
- 2024Learning Memory Kernels in Generalized Langevin Equationscited by 3
- 2023Neural parameter calibration and uncertainty quantification for epidemic forecastingcited by 1
Built on by
Citing papers per year: 2026: 3.
3 papers.