The paper presents a two‑stage method that first trains a neural network on N‑body simulations of multi‑planet systems and then distills its knowledge into closed‑form algebraic equations via symbolic regression. These distilled equations predict the onset of dynamical instability across a wide range of mass ratios, spacings, and eccentricities, outperforming existing analytic estimates. The approach yields a small set of interpretable variables that capture the structure of the instability landscape.
arXiv:2608.22112v1 Announce Type: cross Abstract: We present a machine learning framework for identifying sparse, interpretable models of dynamical systems directly from time-series data. Our approac...
The paper introduces a neural‑network based local‑box chemical kinetics solver for exoplanet atmospheres, employing a residual flow‑map architecture. It achieves microsecond‑scale inference with percent‑level accuracy across a wide range of temperatures, pressures, time steps, and compositional variations, outperforming other machine‑learning models and handling the extreme stiffness of atmospheric chemistry. The surrogate model offers a flexible, efficient alternative to classical solvers for state‑to‑state flow‑map problems in numerical simulations.
arXiv:2608. 02662v1 Announce Type: cross Abstract: Reliable forecasting of nonlinear physical systems underpins scientific discovery and engineering decision-making.
arXiv:2609.24882v1 Announce Type: new Abstract: Hybrid AI-physics climate modeling aims to improve coarse (~100km-resolution) Earth system models by learning to parameterize subgrid processes from hi...
Neural Symbolic Regression (NSR) uses neural networks as functional preconditioners to learn smooth, noise‑robust approximations of target functions in an interaction‑aware nonlinear feature space. A subsequent LASSO step extracts sparse, interpretable closed‑form expressions, while distributed hyperparameter optimization with Ray Tune and ASHA scheduling improves predictive accuracy and symbolic fidelity. Experiments on the Nguyen benchmark demonstrate that NSR outperforms SINDy and untuned neural baselines in RMSE, noise robustness, and out‑of‑distribution generalization, with ablation studies highlighting the importance of feature interactions, neural depth, and tuning strategies.
arXiv:2606. 18464v1 Announce Type: cross Abstract: Detecting the tiny Doppler shifts induced by Earth-mass planets in stellar radial-velocity measurements remains extremely challenging due to stellar activity.
arXiv:2508. 18173v2 Announce Type: replace Abstract: The discovery of symbolic governing equations is a central goal in science; yet, it remains challenging particularly for graph dynamical systems, where the network topology further shapes the system behavior.
arXiv:2607. 27196v1 Announce Type: new Abstract: We present a novel approach to regression tasks using classification which is motivated by the mechanism used by fruitflies to sense their environment.
arXiv:2510. 25306v3 Announce Type: replace Abstract: Partial physical knowledge--governing structures known, constitutive relations or their combinations not--pervades spatiotemporal systems.
arXiv:2607. 00460v1 Announce Type: cross Abstract: Predicting complex spatiotemporal dynamics in physical processes often demands computationally expensive numerical methods or data-driven neural networks that suffer from high training costs, error accumulation, and limited generalizability to unseen parameters.
arXiv:2604. 19465v3 Announce Type: replace-cross Abstract: Understanding how complex systems respond to perturbations, such as whether they will remain stable or what their most sensitive patterns are, is a fundamental challenge across science and engineering.
The article discusses how machine learning can be applied to the Sun’s interior and atmosphere, which are complex, multiscale dynamical systems. It introduces a state-space framework where the Sun’s global state vector is governed by thermodynamic, kinetic, and magnetic properties, and observations provide only partial, projected measurements. The authors propose that machine learning serves as a family of operators—Inference, Emulation, Propagation, and Control—to reconstruct, simulate, and couple this hidden state across different physical layers, thereby bridging observational gaps and accelerating nonlinear simulations to predict the Sun’s global evolution.