Learning to Simulate Unseen Physical Systems with Graph Neural Networks
This addresses the need for efficient and generalizable physical simulations in science and engineering, though it appears incremental by building on existing graph neural network approaches with added physical priors.
The paper tackles the problem of simulating physical systems with neural networks that fail to generalize to unseen materials like liquids with different viscosities, and presents a Graph-based Physics Engine (GPE) that can generalize to materials with different properties not in the training set, performing well in multi-step simulations.
Simulation of the dynamics of physical systems is essential to the development of both science and engineering. Recently there is an increasing interest in learning to simulate the dynamics of physical systems using neural networks. However, existing approaches fail to generalize to physical substances not in the training set, such as liquids with different viscosities or elastomers with different elasticities. Here we present a machine learning method embedded with physical priors and material parameters, which we term as "Graph-based Physics Engine" (GPE), to efficiently model the physical dynamics of different substances in a wide variety of scenarios. We demonstrate that GPE can generalize to materials with different properties not seen in the training set and perform well from single-step predictions to multi-step roll-out simulations. In addition, introducing the law of momentum conservation in the model significantly improves the efficiency and stability of learning, allowing convergence to better models with fewer training steps.