Physics-informed attention-based neural network for solving non-linear partial differential equations

Physics-Informed Neural Networks (PINNs) have enabled significant improvements in modelling physical processes described by partial differential equations (PDEs). PINNs are based on simple architectures, and learn the behavior of complex physical systems by optimizing the network parameters to minimize the residual of the underlying PDE. Current network architectures share some of the limitations of classical numerical discretization schemes when applied to non-linear differential equations in continuum mechanics. A paradigmatic example is the solution of hyperbolic conservation laws that develop highly localized nonlinear shock waves. Learning solutions of PDEs with dominant hyperbolic character is a challenge for current PINN approaches, which rely, like most grid-based numerical schemes, on adding artificial dissipation. Here, we address the fundamental question of which network architectures are best suited to learn the complex behavior of non-linear PDEs. We focus on network architecture rather than on residual regularization. Our new methodology, called Physics-Informed Attention-based Neural Networks, (PIANNs), is a combination of recurrent neural networks and attention mechanisms. The attention mechanism adapts the behavior of the deep neural network to the non-linear features of the solution, and break the current limitations of PINNs. We find that PIANNs effectively capture the shock front in a hyperbolic model problem, and are capable of providing high-quality solutions inside and beyond the training set.

[1]  Ramakrishna Tipireddy,et al.  Conditional Karhunen-Loève expansion for uncertainty quantification and active learning in partial differential equation models , 2019, J. Comput. Phys..

[2]  Teeratorn Kadeethum,et al.  Physics-informed neural networks for solving nonlinear diffusivity and Biot’s equations , 2020, PloS one.

[3]  George Em Karniadakis,et al.  NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations , 2020, J. Comput. Phys..

[4]  H. Tchelepi,et al.  LIMITATIONS OF PHYSICS INFORMED MACHINE LEARNING FOR NONLINEAR TWO-PHASE TRANSPORT IN POROUS MEDIA , 2020 .

[5]  G. Beroza,et al.  Laboratory earthquake forecasting: A machine learning competition , 2021, Proceedings of the National Academy of Sciences.

[6]  M. Raissi,et al.  A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics , 2021, Computer Methods in Applied Mechanics and Engineering.

[7]  Timon Rabczuk,et al.  An Energy Approach to the Solution of Partial Differential Equations in Computational Mechanics via Machine Learning: Concepts, Implementation and Applications , 2019, Computer Methods in Applied Mechanics and Engineering.

[8]  Paris Perdikaris,et al.  Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations , 2019, J. Comput. Phys..

[9]  S. E. Buckley,et al.  Mechanism of Fluid Displacement in Sands , 1942 .

[10]  Yoshua Bengio,et al.  Neural Machine Translation by Jointly Learning to Align and Translate , 2014, ICLR.

[11]  George Em Karniadakis,et al.  Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations , 2020, Science.

[12]  Liu Yang,et al.  B-PINNs: Bayesian Physics-Informed Neural Networks for Forward and Inverse PDE Problems with Noisy Data , 2020, J. Comput. Phys..

[13]  Randall J. LeVeque,et al.  Numerical methods for conservation laws (2. ed.) , 1992, Lectures in mathematics.

[14]  C. Dafermos Hyberbolic Conservation Laws in Continuum Physics , 2000 .

[15]  Hamdi A. Tchelepi,et al.  Physics Informed Deep Learning for Transport in Porous Media. Buckley Leverett Problem , 2020, ArXiv.

[16]  Quoc V. Le,et al.  Sequence to Sequence Learning with Neural Networks , 2014, NIPS.

[17]  Stephan Hoyer,et al.  Learning data-driven discretizations for partial differential equations , 2018, Proceedings of the National Academy of Sciences.

[18]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[19]  G. Karniadakis,et al.  Physics-informed neural networks for high-speed flows , 2020, Computer Methods in Applied Mechanics and Engineering.

[20]  H. Tchelepi,et al.  Physics Informed Deep Learning for Flow and Transport in Porous Media , 2021, Day 1 Tue, October 26, 2021.

[21]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[22]  Arnulf Jentzen,et al.  Solving high-dimensional partial differential equations using deep learning , 2017, Proceedings of the National Academy of Sciences.

[23]  Julian Togelius,et al.  Deep Reinforcement Learning for General Video Game AI , 2018, 2018 IEEE Conference on Computational Intelligence and Games (CIG).

[24]  George Em Karniadakis,et al.  Quantifying total uncertainty in physics-informed neural networks for solving forward and inverse stochastic problems , 2018, J. Comput. Phys..

[25]  Petros Koumoutsakos,et al.  Machine Learning for Fluid Mechanics , 2019, Annual Review of Fluid Mechanics.

[26]  Paris Perdikaris,et al.  Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations , 2017, ArXiv.

[27]  Yoshua Bengio,et al.  Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation , 2014, EMNLP.

[28]  Terrence J Sejnowski,et al.  The unreasonable effectiveness of deep learning in artificial intelligence , 2020, Proceedings of the National Academy of Sciences.

[29]  Christian Beck,et al.  An overview on deep learning-based approximation methods for partial differential equations , 2020, ArXiv.

[30]  G. Karniadakis,et al.  Conservative physics-informed neural networks on discrete domains for conservation laws: Applications to forward and inverse problems , 2020 .

[31]  Julian Togelius,et al.  Bootstrapping Conditional GANs for Video Game Level Generation , 2019, 2020 IEEE Conference on Games (CoG).

[32]  Todd A. Oliver,et al.  Solving differential equations using deep neural networks , 2020, Neurocomputing.

[33]  Sorin Grigorescu,et al.  A Survey of Deep Learning Techniques for Autonomous Driving , 2020, J. Field Robotics.

[34]  Alfio Quarteroni,et al.  Machine learning for fast and reliable solution of time-dependent differential equations , 2019, J. Comput. Phys..