An Energy-Based Conservative-Dissipative Latent Neural Evolution Operator for Magnetization Dynamics
Abstract
We introduce a reduced-order learning framework that couples a convolutional autoencoder to a structured latent neural ordinary differential equation. A learned scalar generates the latent dynamics through an antisymmetric operator and a positive-definite dissipative operator. The antisymmetric component permits motion along scalar level sets, while the dissipative component decreases the scalar. Adding a coercive quadratic term gives bounded continuous-time latent trajectories and approach to the scalar's critical set under the stated assumptions. The scalar provides a structural constraint rather than an approximation of physical energy. The encoder, decoder, scalar, and operators are trained jointly on short snapshot windows without time-derivative supervision, physical-energy labels, or dissipation penalties. At inference, one initial state is encoded and evolved continuously in latent space, with decoding only at requested times. We demonstrate the framework on micromagnetic magnetization trajectories from NIST μMAG Standard Problem 4 and compare it with dissipative-only dynamics and an unstructured latent neural differential equation. Repeated training runs show competitive observed-horizon accuracy, while selected longer rollouts illustrate improved retention of approximately stationary states relative to the unstructured baseline. These results motivate scalar-generated latent dynamics as an inductive bias for reduced-order prediction and other learning tasks where bounded evolution and equilibrium behavior are important.
Related
- CLNO: Conservative Local Neural Operator for Compressible Flow Prediction
- Dissipative Hamiltonian representations of linear evolution equations
- The mean-field limit of non-exchangeable particle systems with non-conservative dynamics and adaptive weights
- Transport Polynomial Exactness for Direct ALE-DG Methods with Nonsmooth mesh motion: A Conservative Coefficient-Evolution Framework
- A characteristic mapping method for dissipative evolution equations
BibTeX
@misc{schaffer2026energybased,
title = {{An Energy-Based Conservative-Dissipative Latent Neural Evolution Operator for Magnetization Dynamics}},
author = {Sebastian Schaffer and Lukas Exl},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2609.04530},
eprint = {2609.04530},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.04530},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - An Energy-Based Conservative-Dissipative Latent Neural Evolution Operator for Magnetization Dynamics
AU - Sebastian Schaffer
AU - Lukas Exl
PY - 2026
DA - 2026/10/24
PB - arXiv
DO - 10.48550/arxiv.2609.04530
UR - https://arxiv.org/abs/2609.04530
AB - We introduce a reduced-order learning framework that couples a convolutional autoencoder to a structured latent neural ordinary differential equation. A learned scalar generates the latent dynamics through an antisymmetric operator and a positive-definite dissipative operator. The antisymmetric component permits motion along scalar level sets, while the dissipative component decreases the scalar. Adding a coercive quadratic term gives bounded continuous-time latent trajectories and approach to the scalar's critical set under the stated assumptions. The scalar provides a structural constraint rather than an approximation of physical energy. The encoder, decoder, scalar, and operators are trained jointly on short snapshot windows without time-derivative supervision, physical-energy labels, or dissipation penalties. At inference, one initial state is encoded and evolved continuously in latent space, with decoding only at requested times. We demonstrate the framework on micromagnetic magnetization trajectories from NIST μMAG Standard Problem 4 and compare it with dissipative-only dynamics and an unstructured latent neural differential equation. Repeated training runs show competitive observed-horizon accuracy, while selected longer rollouts illustrate improved retention of approximately stationary states relative to the unstructured baseline. These results motivate scalar-generated latent dynamics as an inductive bias for reduced-order prediction and other learning tasks where bounded evolution and equilibrium behavior are important.
ER -