REVIEW 2 cited by
Investigating and Mitigating Failure Modes in Physics-informed Neural Networks (PINNs)
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Investigating and Mitigating Failure Modes in Physics-informed Neural Networks (PINNs)
read the original abstract
This paper explores the difficulties in solving partial differential equations (PDEs) using physics-informed neural networks (PINNs). PINNs use physics as a regularization term in the objective function. However, a drawback of this approach is the requirement for manual hyperparameter tuning, making it impractical in the absence of validation data or prior knowledge of the solution. Our investigations of the loss landscapes and backpropagated gradients in the presence of physics reveal that existing methods produce non-convex loss landscapes that are hard to navigate. Our findings demonstrate that high-order PDEs contaminate backpropagated gradients and hinder convergence. To address these challenges, we introduce a novel method that bypasses the calculation of high-order derivative operators and mitigates the contamination of backpropagated gradients. Consequently, we reduce the dimension of the search space and make learning PDEs with non-smooth solutions feasible. Our method also provides a mechanism to focus on complex regions of the domain. Besides, we present a dual unconstrained formulation based on Lagrange multiplier method to enforce equality constraints on the model's prediction, with adaptive and independent learning rates inspired by adaptive subgradient methods. We apply our approach to solve various linear and non-linear PDEs.
Forward citations
Cited by 2 Pith papers
-
A physics-informed neural network approach to the point defect model for electrochemical oxide film growth
A hybrid PINN anchored by one FEM data point reproduces point-defect-model film thicknesses to about 1% error, while the pure PINN overpredicts by 2,400-5,700%.
-
Physics-Informed Neural Networks in Clean Combustion: A Pathway to Sustainable Aerospace Propulsion
A review article argues physics-informed neural networks are a faster, more data-efficient route to clean combustion modeling, but its 'transformative' thesis is undercut by its own scaling caveats and duplicated sections.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.