Corpus ID: 231583090

A Bayesian neural network predicts the dissolution of compact planetary systems

@article{Cranmer2021ABN,
  title={A Bayesian neural network predicts the dissolution of compact planetary systems},
  author={M. Cranmer and D. Tamayo and H. Rein and P. Battaglia and S. Hadden and P. Armitage and S. Ho and D. N. Spergel},
  journal={ArXiv},
  year={2021},
  volume={abs/2101.04117}
}
  • M. Cranmer, D. Tamayo, +5 authors D. N. Spergel
  • Published 2021
  • Computer Science, Physics, Mathematics
  • ArXiv
  • Despite over three hundred years of effort, no solutions exist for predicting when a general planetary configuration will become unstable. We introduce a deep learning architecture to push forward this problem for compact systems. While current machine learning algorithms in this area rely on scientist-derived instability metrics, our new technique learns its own metrics from scratch, enabled by a novel internal structure inspired from dynamics theory. Our Bayesian neural network model can… CONTINUE READING

    Figures and Tables from this paper

    References

    SHOWING 1-10 OF 84 REFERENCES
    A Machine Learns to Predict the Stability of Tightly Packed Planetary Systems
    • 29
    • PDF
    Predicting the long-term stability of compact multiplanet systems
    • 4
    • PDF
    Fundamental limits from chaos on instability time predictions in compact planetary systems
    • 5
    • PDF
    Bayesian Deep Learning and a Probabilistic Perspective of Generalization
    • 44
    • Highly Influential
    • PDF
    Dynamical instability and its implications for planetary system architecture
    • 14
    • PDF
    Nekhoroshev Estimates for the Survival Time of Tightly Packed Planetary Systems
    • 5
    • PDF
    The Case for Bayesian Deep Learning
    • A. Wilson
    • Computer Science, Mathematics
    • ArXiv
    • 2020
    • 25
    • PDF
    Learning Symbolic Physics with Graph Networks
    • 17
    • PDF