Approximating the solution to wave propagation using deep neural networks

Abstract

Humans gain an implicit understanding of physical laws through observing and interacting with the world. Endowing an autonomous agent with an understanding of physical laws through experience and observation is seldom practical: we should seek alternatives. Fortunately, many of the laws of behaviour of the physical world can be derived from prior knowledge of dynamical systems, expressed through the use of partial differential equations. In this work, we suggest a neural network capable of understanding a specific physical phenomenon: wave propagation in a two-dimensional medium. We define `understanding' in this context as the ability to predict the future evolution of the spatial patterns of rendered wave amplitude from a relatively small set of initial observations. The inherent complexity of the wave equations -- together with the existence of reflections and interference -- makes the prediction problem non-trivial. A network capable of making approximate predictions also unlocks the opportunity to speed-up numerical simulations for wave propagation. To this aim, we created a novel dataset of simulated wave motion and built a predictive deep neural network comprising of three main blocks: an encoder, a propagator made by 3 LSTMs, and a decoder. Results show reasonable predictions for as long as 80 time steps into the future on a dataset not seen during training. Furthermore, the network is able to generalize to an initial condition that is qualitatively different from those seen during training.

arXiv commentAccepted to the NeurIPS 2018 Workshop "Modeling the Physical World: Perception, Learning, and Control" (extended abstract)

References in corpus (5)

• Using Machine Learning to Replicate Chaotic Attractors and Calculate Lyapunov Exponents from Data
• Graph networks as learnable physics engines for inference and control
• Unsupervised Learning for Physical Interaction through Video Prediction
• Learning A Physical Long-term Predictor
• SPNets: Differentiable Fluid Dynamics for Deep Neural Networks

Cited by in corpus (6)

• PhyCRNet: Physics-informed Convolutional-Recurrent Network for Solving Spatiotemporal PDEs
• Physics-informed neural wavefields with Gabor basis functions
• Prediction of Ultrasonic Guided Wave Propagation in Solid-fluid and their Interface under Uncertainty using Machine Learning
• Simulating Continuum Mechanics with Multi-Scale Graph Neural Networks
• Solving Image PDEs with a Shallow Network
• Wavefield solutions from machine learned functions