A Bayesian neural network predicts the dissolution of compact planetary systems
This work provides a significantly more accurate and faster method for predicting the long-term stability of compact planetary systems, which is a long-standing problem for astrophysicists and planetary scientists.
This paper introduces a Bayesian neural network that predicts the instability and dissolution time of compact planetary systems. The model achieves over two orders of magnitude higher accuracy than analytical estimators and reduces bias by nearly a factor of three compared to existing machine learning algorithms, while also being five orders of magnitude faster than numerical integrators.
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 accurately predict not only if, but also when a compact planetary system with three or more planets will go unstable. Our model, trained directly from short N-body time series of raw orbital elements, is more than two orders of magnitude more accurate at predicting instability times than analytical estimators, while also reducing the bias of existing machine learning algorithms by nearly a factor of three. Despite being trained on compact resonant and near-resonant three-planet configurations, the model demonstrates robust generalization to both non-resonant and higher multiplicity configurations, in the latter case outperforming models fit to that specific set of integrations. The model computes instability estimates up to five orders of magnitude faster than a numerical integrator, and unlike previous efforts provides confidence intervals on its predictions. Our inference model is publicly available in the SPOCK package, with training code open-sourced.