Efficient Learning of Accurate Surrogates for Simulations of Complex Systems
arXiv:2207.12855 · doi:10.1038/s42256-024-00839-1
Abstract
Machine learning methods are increasingly used to build computationally inexpensive surrogates for complex physical models. The predictive capability of these surrogates suffers when data are noisy, sparse, or time-dependent. As we are interested in finding a surrogate that provides valid predictions of any potential future model evaluations, we introduce an online learning method empowered by optimizer-driven sampling. The method has two advantages over current approaches. First, it ensures that all turning points on the model response surface are included in the training data. Second, after any new model evaluations, surrogates are tested and "retrained" (updated) if the "score" drops below a validity threshold. Tests on benchmark functions reveal that optimizer-directed sampling generally outperforms traditional sampling methods in terms of accuracy around local extrema, even when the scoring metric favors overall accuracy. We apply our method to simulations of nuclear matter to demonstrate that highly accurate surrogates for the nuclear equation of state can be reliably auto-generated from expensive calculations using a few model evaluations.
13 pages, 6 figures, submitted to Nature Machine Intelligence
References in corpus (5)
- GW170817: Observation of Gravitational Waves from a Binary Neutron Star Inspiral
- PSR J0030+0451 Mass and Radius from NICER Data and Implications for the Properties of Neutron Star Matter
- A NICER View of PSR J0030+0451: Millisecond Pulsar Parameter Estimation
- Path Integral Monte Carlo Simulation of the Warm-Dense Homogeneous Electron Gas
- Finite-temperature extension for cold neutron star equations of state