GENERIC-FNO: Embedding Energy Conservation and Entropy Production into Fourier Neural Operators

• URL: http://arxiv.org/abs/2606.08343v2
• Date: Wed, 10 Jun 2026 13:42:58 GMT
• Title: GENERIC-FNO: Embedding Energy Conservation and Entropy Production into Fourier Neural Operators
• Abstract summary: We introduce GENERIC-FNO, the first neural operator to embed the full GENERIC structure of nonequilibrium thermodynamics directly in function space.<n>The continuous-time dynamics conserve the learned energy and produce entropy exactly.<n>Across three operator backbones (1D/2D FNOs and DeepONet) and four PDEs spanning reversible, dissipative, and mixed regimes, GENERIC-FNO preserves its exact structural guarantees zero-shot across a 4x super-resolution range (64 to 256)
• Score: 1.037435491622876
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We introduce GENERIC-FNO, the first neural operator to embed the full GENERIC (metriplectic) structure of nonequilibrium thermodynamics -- reversible, energy-conserving dynamics and irreversible, entropy-producing dynamics coupled through the degeneracy conditions -- directly in function space. Existing structure-preserving neural operators enforce at most a single conservation law or reversible (Hamiltonian) structure, while thermodynamically consistent learning has been confined to finite-dimensional, graph, or particle systems. GENERIC-FNO closes this gap: it learns the energy and entropy functionals as neural operators and parameterizes the Poisson and friction operators as diagonal Fourier multipliers sandwiched between rank-one projections that enforce the degeneracy conditions exactly, by construction, with no penalty term, update projection, or residual. The degeneracy identities hold to machine precision (residuals ~10^-13) for any initialization, dimension, or resolution, so the continuous-time dynamics conserve the learned energy and produce entropy exactly; the explicit time stepping adds only a small O(dt^2) drift (per-step residual ~10^-6). We further note that the (E,S,L,M) decomposition of a given flow is not unique, and introduce a gauge-invariant dissipation diagnostic separating reversible from dissipative dynamics independently of the learned functionals. Across three operator backbones (1D/2D FNOs and DeepONet) and four PDEs spanning reversible, dissipative, and mixed regimes, GENERIC-FNO preserves its exact structural guarantees zero-shot across a 4x super-resolution range (64 to 256), recovers the ground-truth ordering of physical dissipation, and is competitive with strong unconstrained and energy-penalized baselines, outperforming them on several dissipative and mixed problems at comparable or fewer parameters.

Related papers

• An Energy-Based Conservative-Dissipative Latent Neural Evolution Operator for Magnetization Dynamics [0.0]
  We develop an energy-based reduced-order model for micromagnetic magnetization dynamics.<n>Motivated by the precessional-dissipative structure of the Landau-Lifshitz-Gilbert equation, the latent vector field is generated.<n>Dissipative-only and antisymmetric-dissipative models achieve comparable accuracy on short training-style windows.<n>The deep-quadratic energy gives the best overall accuracy for both field directions.
  arXiv  Detail & Related papers  (2026-09-03T22:42:04Z)
• Geometry-aware Incremental Neural Operator for Long-Horizon PDE prediction [17.3214708475049]
  We propose a geometry-aware incremental neural operator (GeoIncNO) for stable long-horizon PDE prediction.<n>GeoIncNO predicts latent increments for residual advancement and uses lightweight low-rank projectors to regulate channel coupling within active frequency bands.<n>Experiments on six PDE benchmarks, covering 1D, 2D, and 3D dynamical systems, show that GeoIncNO consistently achieves strong prediction accuracy, improved rollout stability, and better spectral fidelity compared with competitive neural-operator baselines.
  arXiv  Detail & Related papers  (2026-07-31T07:46:35Z)
• Towards more accurate natural orbital functional approximations: including 4-index cumulant contributions [0.0]
  We revisit the construction of natural orbital functionals by correcting the cumulant contribution produced by the PNOF5 functional.<n>Our method enforces known contributions of the cumulant to local spin fragments and purify the delocalization index at the dissociation limit.<n>The resulting functional yields improved behavior in strongly correlated regimes.
  arXiv  Detail & Related papers  (2026-07-23T14:59:50Z)
• Energy-Structured Low-Rank Adaptation for Continual Learning [78.5390284258307]
  We propose textbfEnergy-Concentrated and textbfEnergy-Ordered textbfLow-textbfRank textbfAdaptation (E$2$-LoRA)<n>By explicitly ordering and concentrating knowledge into leading ranks, E$2$-LoRA frees capacity for subsequent tasks.
  arXiv  Detail & Related papers  (2026-05-26T12:55:02Z)
• Non-Associativity Induced Modifications of Open-System Quantum Dynamics: General Master Equation and a Two-Qubit Ising Case Study [0.0]
  Nonassociative deformations of phase-space structures arise naturally in the presence of magnetic charge.<n>We show that weak nonassociativity manifests as a coherent, population-dependent deformation of open-system dynamics.
  arXiv  Detail & Related papers  (2026-04-17T18:26:39Z)
• Learning noisy phase transition dynamics from stochastic partial differential equations [5.243031075965086]
  Machine-learning surrogates must represent nuity explicitly, enforce conservation laws by construction, and expose physically interpretable structure.<n>We develop physics-independent surrogate models for the Cahn-Hilliard equation in 3D that satisfy all three requirements simultaneously.<n>A learnable free energy functional provides thermodynamic interpretability, validated by independent recovery of the bulk double-well landscape.
  arXiv  Detail & Related papers  (2026-03-31T02:43:38Z)
• An SO(3)-equivariant reciprocal-space neural potential for long-range interactions [84.08721947253646]
  Long-range electrostatic and polarization interactions play a central role in molecular and condensed-phase systems.<n>Modern SO(3)-equivariant neural potentials achieve high accuracy for short-range chemistry, but cannot represent the anisotropic, slowly decaying multipolar correlations governing realistic materials.<n>Here we introduce EquiEwald, a unified neural interatomic potential that embeds an Ewald-inspired reciprocal-space formulation within an irreducible SO(3)-equivariant framework.
  arXiv  Detail & Related papers  (2026-03-19T01:16:11Z)
• Manifold-Orthogonal Dual-spectrum Extrapolation for Parameterized Physics-Informed Neural Networks [5.601252185916962]
  Physics-informed neural networks (PINNs) have achieved notable success in modeling dynamical systems governed by partial differential equations (PDEs)<n>We propose Manifold-Orthogonal Dual-spectrum Extrapolation (MODE), a lightweight micro-architecture designed for physics operator adaptation.
  arXiv  Detail & Related papers  (2026-03-14T04:52:50Z)
• Structure-Preserving Physics-Informed Neural Network for the Korteweg--de Vries (KdV) Equation [0.0]
  This paper introduces a emphstructure-preserving PINN framework for the nonlinear Korteweg--de Vries (KdV) equation.<n>The proposed method embeds the conservation of mass and Hamiltonian energy directly into the loss function.<n>We successfully reproduces hallmark behaviors of KdV dynamics while maintaining conserved invariants.
  arXiv  Detail & Related papers  (2025-11-01T06:07:24Z)
• Differential-Integral Neural Operator for Long-Term Turbulence Forecasting [43.04613533979613]
  textbfunderlineDifferential-textbfunderlineIntegral textbfunderlineNeural textbfunderlineOperator (method)<n>method explicitly models the turbulent evolution through parallel branches that learn distinct physical operators.<n>It successfully suppresses error accumulation over hundreds of timesteps, maintains high fidelity in both the vorticity fields and energy spectra, and establishes a new benchmark for physically consistent, long-range turbulence forecast.
  arXiv  Detail & Related papers  (2025-09-25T14:08:26Z)
• Metriplectic Conditional Flow Matching for Dissipative Dynamics [5.920407670799846]
  conditional flow matching learns dissipative dynamics without violating first principles.<n>MCFM trains via conditional flow matching on short transitions, avoiding long rollout adjoints.<n>We provide continuous and discrete time guarantees linking this parameterization and sampler to conservation, monotonic dissipation, and stable rollouts.
  arXiv  Detail & Related papers  (2025-09-23T19:46:54Z)
• TANGO: Time-Reversal Latent GraphODE for Multi-Agent Dynamical Systems [43.39754726042369]
  We propose a simple-yet-effective self-supervised regularization term as a soft constraint that aligns the forward and backward trajectories predicted by a continuous graph neural network-based ordinary differential equation (GraphODE) It effectively imposes time-reversal symmetry to enable more accurate model predictions across a wider range of dynamical systems under classical mechanics. Experimental results on a variety of physical systems demonstrate the effectiveness of our proposed method.
  arXiv  Detail & Related papers  (2023-10-10T08:52:16Z)
• Message-Passing Neural Quantum States for the Homogeneous Electron Gas [41.94295877935867]
  We introduce a message-passing-neural-network-based wave function Ansatz to simulate extended, strongly interacting fermions in continuous space. We demonstrate its accuracy by simulating the ground state of the homogeneous electron gas in three spatial dimensions.
  arXiv  Detail & Related papers  (2023-05-12T04:12:04Z)
• Probing eigenstate thermalization in quantum simulators via fluctuation-dissipation relations [77.34726150561087]
  The eigenstate thermalization hypothesis (ETH) offers a universal mechanism for the approach to equilibrium of closed quantum many-body systems. Here, we propose a theory-independent route to probe the full ETH in quantum simulators by observing the emergence of fluctuation-dissipation relations. Our work presents a theory-independent way to characterize thermalization in quantum simulators and paves the way to quantum simulate condensed matter pump-probe experiments.
  arXiv  Detail & Related papers  (2020-07-20T18:00:02Z)

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.