Large-step neural network for learning the symplectic evolution from partitioned data
arXiv:2208.14148 · doi:10.1093/mnras/stad1948
Abstract
In this study, we focus on learning Hamiltonian systems, which involves predicting the coordinate (q) and momentum (p) variables generated by a symplectic mapping. Based on Chen & Tao (2021), the symplectic mapping is represented by a generating function. To extend the prediction time period, we develop a new learning scheme by splitting the time series (q_i, p_i) into several partitions. We then train a large-step neural network (LSNN) to approximate the generating function between the first partition (i.e. the initial condition) and each one of the remaining partitions. This partition approach makes our LSNN effectively suppress the accumulative error when predicting the system evolution. Then we train the LSNN to learn the motions of the 2:3 resonant Kuiper belt objects for a long time period of 25000 yr. The results show that there are two significant improvements over the neural network constructed in our previous work (Li et al. 2022): (1) the conservation of the Jacobi integral, and (2) the highly accurate predictions of the orbital evolution. Overall, we propose that the designed LSNN has the potential to considerably improve predictions of the long-term evolution of more general Hamiltonian systems.
arXiv comment13 pages, 7 figures, accepted for publication in MNRAS
References in corpus (20)
- THE ORIGIN OF PLUTO'S ORBIT: IMPLICATIONS FOR THE SOLAR SYSTEM BEYOND NEPTUNE
- Neptune's Migration into a Stirred-Up Kuiper Belt: A Detailed Comparison of Simulations to Observations
- Neptune's Orbital Migration Was Grainy, Not Smooth
- Explicit symplectic approximation of nonseparable Hamiltonians: algorithm and long time performance
- Newton vs the machine: solving the chaotic three-body problem using deep neural networks
- Painting galaxies into dark matter halos using machine learning
- Machine-learning Approaches to Exoplanet Transit Detection and Candidate Validation in Wide-field Ground-based Surveys
- The miniJPAS survey: star-galaxy classification using machine learning
- Machine learning technique for morphological classification of galaxies from the SDSS. I. Photometry-based approach
- An interpretable machine learning framework for dark matter halo formation
- Exoplanet Validation with Machine Learning: 50 new validated Kepler planets
- OSSOS: XIII. Fossilized Resonant Dropouts Tentatively Confirm Neptune's Migration was Grainy and Slow
- Machine Learning applied to asteroid dynamics
- Structure-preserving Method for Reconstructing Unknown Hamiltonian Systems from Trajectory Data
- A study of the high-inclination population in the Kuiper belt -- 1. The Plutinos
- Details of Resonant Structures Within a Nice Model Kuiper Belt: Predictions for High-Perihelion TNO Detections
- Machine Learning Classifiers for Intermediate Redshift Emission Line Galaxies
- Machine learning prediction for mean motion resonance behaviour -- The planar case
- Machine Learning the Fates of Dark Matter Subhalos: A Fuzzy Crystal Ball
- A study of the high-inclination population in the Kuiper belt -- IV. High-order mean motion resonances in the classical region