Learning in Sinusoidal Spaces with Physics-Informed Neural Networks

Abstract

A physics-informed neural network (PINN) uses physics-augmented loss functions, e.g., incorporating the residual term from governing partial differential equations (PDEs), to ensure its output is consistent with fundamental physics laws. However, it turns out to be difficult to train an accurate PINN model for many problems in practice. In this paper, we present a novel perspective of the merits of learning in sinusoidal spaces with PINNs. By analyzing behavior at model initialization, we first show that a PINN of increasing expressiveness induces an initial bias around flat output functions. Notably, this initial solution can be very close to satisfying many physics PDEs, i.e., falling into a local minimum of the PINN loss that only minimizes PDE residuals, while still being far from the true solution that jointly minimizes PDE residuals and the initial and/or boundary conditions. It is difficult for gradient descent optimization to escape from such a local minimum trap, often causing the training to stall. We then prove that the sinusoidal mapping of inputs, in an architecture we label as sf-PINN, is effective to increase input gradient variability, thus avoiding being trapped in such deceptive local minimum. The level of variability can be effectively modulated to match high-frequency patterns in the problem at hand. A key facet of this paper is the comprehensive empirical study that demonstrates the efficacy of learning in sinusoidal spaces with PINNs for a wide range of forward and inverse modelling problems spanning multiple physics domains.

16 pages, 13 figures

References in corpus (4)

• Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains
• CAN-PINN: A Fast Physics-Informed Neural Network Based on Coupled-Automatic-Numerical Differentiation Method
• Adversarial Multi-task Learning Enhanced Physics-informed Neural Networks for Solving Partial Differential Equations
• Training multi-objective/multi-task collocation physics-informed neural network with student/teachers transfer learnings

Cited by in corpus (13)

• CAN-PINN: A Fast Physics-Informed Neural Network Based on Coupled-Automatic-Numerical Differentiation Method
• Physics-informed neural networks of the Saint-Venant equations for downscaling a large-scale river model
• Data vs. Physics: The Apparent Pareto Front of Physics-Informed Neural Networks
• Learning Specialized Activation Functions for Physics-informed Neural Networks
• Improved Training of Physics-Informed Neural Networks with Model Ensembles
• Noise-aware Physics-informed Machine Learning for Robust PDE Discovery
• Point Neuron Learning: A New Physics-Informed Neural Network Architecture
• Reconstructing unsteady flows from sparse, noisy measurements with a physics-constrained convolutional neural network
• Decoder Decomposition for the Analysis of the Latent Space of Nonlinear Autoencoders With Wind-Tunnel Experimental Data
• Forward and Inverse Simulation of Pseudo-Two-Dimensional Model of Lithium-Ion Batteries Using Neural Networks
• LSA-PINN: Linear Boundary Connectivity Loss for Solving PDEs on Complex Geometry
• Evolutionary Optimization of Physics-Informed Neural Networks: Evo-PINN Frontiers and Opportunities
• REAct: Rational Exponential Activation for Better Learning and Generalization in PINNs