arXiv:2607. 23501v1 Announce Type: new Abstract: Locating periodic solutions of chaotic dynamical systems normally requires an initial guess close enough to the target orbit for numerical continuation or gradient-based search to converge.
arXiv:2607. 02194v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) have emerged as a promising route to solve partial differential equations, yet they have struggled to reach the precision of classical solvers.
arXiv:2607. 20235v1 Announce Type: cross Abstract: Classical regularization removes the binary-collision singularity from the Kepler problem, but its value as a representation for learned Hamiltonian dynamics has not been systematically isolated.
The paper introduces SINFONIA, a family of neural-flow architectures designed to learn finite-time evolution maps for orbital dynamics while preserving key physical structures. Three variants—symplectic/slimplectic, Taylor-anchored, and Magnusian—are applied to a 2.5PN neutron-star inspiral, demonstrating that long-term accuracy depends on a single secular channel tied to energy–angular-momentum balance rather than pointwise errors. These learned maps achieve accurate phase evolution over up to $10^{5}$ orbital periods at lower computational cost than traditional integrators, and can also be used to infer unmodeled forces from accumulated phase data.
arXiv:2606. 09929v1 Announce Type: cross Abstract: Physical reservoir computing harnesses nonlinear mechanical dynamics but, by convention, freezes the substrate and trains only a linear readout, presuming the substrate is not usefully trainable.
The paper introduces SINFONIA, a family of neural‑flow architectures designed to learn structure‑preserving evolution maps for long‑duration gravitational‑wave modelling. Three variants—symplectic, slimplectic, and Magnusian—are trained on a 2.5PN neutron‑star inspiral and demonstrate that long‑time accuracy is governed by a single secular channel linked to energy–angular‑momentum balance, allowing accurate integration over up to $10^{5}$ orbital periods with far fewer computational steps than traditional integrators. The learned maps also enable physics inference, recovering un‑modelled dissipative forces from accumulated phase information.
Autoregressive transformers trained on limited trajectories of nonlinear dynamical systems can extrapolate to unseen parameter regimes, reproducing period-doubling cascades, chaotic dynamics, and attractor structures with high fidelity. In the logistic map, the model captures successive period doublings up to period 128, achieving a scaling ratio within $5 imes10^{-4}$ of the Feigenbaum constant. The study also shows how control‑parameter information is processed via attention, shaping the closed‑loop dynamics during training.
The paper introduces a world model that learns to predict the evolution of physical systems while respecting key physical principles. By hard‑coding a general structure—generating dynamics from the gradient of a learned energy via a fixed reversible operator and imposing constraints on energy, dissipation, and interventions—the model achieves second‑law compatible dissipation, accurate responses to parameter changes, long‑term stability, and robustness to disturbances. Experiments on an electromagnetic cavity, a particle‑in‑cell grid, and shallow‑water fluid demonstrate that the model can recover accurate constitutive functions, distinguish conserving from dissipating regimes, and transfer learned physics to unseen conditions, outperforming unconstrained models.
arXiv:2608. 08322v1 Announce Type: cross Abstract: Physics-informed neural networks applied to the level-set formulation of interface advection commonly augment the residual and initial-condition losses with an eikonal regulariser, penalising the deviation of $\|\nabla\phi\|$ from unity.
arXiv:2507. 03631v4 Announce Type: replace Abstract: Extracting interpretable mathematical models from complex dynamical systems is difficult, especially for chaotic dynamics observed with noisy experimental data.
arXiv:2606. 30789v1 Announce Type: new Abstract: Group Relative Policy Optimization (GRPO) has become a standard tool for improving the reasoning ability of large language models, yet its training dynamics are still described empirically: reward trajectories are fit with low-parameter functional forms whose constants carry no mechanistic meaning, and hyperparameter choices remain a matter of trial and error.
arXiv:2609.24947v1 Announce Type: new Abstract: Neural operators evaluate parametric partial differential equations cheaply but degrade sharply outside their training distribution. Physics-informed n...
arXiv:2608. 10235v1 Announce Type: new Abstract: Hamiltonian Neural Networks (HNNs) parameterize conservative dynamics through a learned scalar Hamiltonian, providing an architectural prior that is absent from generic vector-field neural networks.