Abstract:Physics-based models are widely used to study dynamical systems in a variety of scientific and engineering problems. However, these models are necessarily approximations of reality due to incomplete knowledge or excessive complexity in modeling underlying processes. As a result, they often produce biased simulations due to inaccurate parameterizations or approximations used to represent the true physics. In this paper, we aim to build a new physics-guided machine learning framework to monitor dynamical systems. The idea is to use advanced machine learning model to extract complex spatio-temporal data patterns while also incorporating general scientific knowledge embodied in simulated data generated by the physics-based model. To handle the bias in simulated data caused by imperfect parameterization, we propose to extract general physical relations jointly from multiple sets of simulations generated by a physics-based model under different physical parameters. In particular, we develop a spatio-temporal network architecture that uses its gating variables to capture the variation of physical parameters. We initialize this model using a pre-training strategy that helps discover common physical patterns shared by different sets of simulated data. Then, we fine-tune it combining limited observations and adequate simulations. By leveraging the complementary strength of machine learning and domain knowledge, our method has been shown to produce accurate predictions, use less training samples and generalize to out-of-sample scenarios. We further show that the method can provide insights about the variation of physical parameters over space and time in two domain applications: predicting temperature in streams and predicting temperature in lakes.

Learning force laws in many-body systems

Discovery of Physics from Data: Universal Laws and Discrepancies

Integrating Machine Learning with Physics-Based Modeling

Physics-guided machine learning from simulated data with different physical parameters

Discovering dynamic laws from observations: the case of self-propelled, interacting colloids

Machine learning for many-body physics: The case of the Anderson impurity model

Accurate machine learning force fields via experimental and simulation data fusion

Structural dynamics simulation using a novel physics-guided machine learning method

Learning about learning by many-body systems

Machine learning and the physical sciences

Electronic Excited States from Physically Constrained Machine Learning

Electronic excited states from physically-constrained machine learning

Discovering the Building Blocks of Atomic Systems using Machine Learning

The dark side of the forces: assessing non-conservative force models for atomistic machine learning

Complexity of many-body interactions in transition metals via machine-learned force fields from the TM23 data set

Synergistic Integration of Physical Embedding and Machine Learning Enabling Precise and Reliable Force Field

Machine Learning Interatomic Potentials and Long-Range Physics

Physics-informed machine learning