Characterizing possible failure modes in physics-informed neural networks

Abstract

Recent work in scientific machine learning has developed so-called physics-informed neural network (PINN) models. The typical approach is to incorporate physical domain knowledge as soft constraints on an empirical loss function and use existing machine learning methodologies to train the model. We demonstrate that, while existing PINN methodologies can learn good models for relatively trivial problems, they can easily fail to learn relevant physical phenomena for even slightly more complex problems. In particular, we analyze several distinct situations of widespread physical interest, including learning differential equations with convection, reaction, and diffusion operators. We provide evidence that the soft regularization in PINNs, which involves PDE-based differential operators, can introduce a number of subtle problems, including making the problem more ill-conditioned. Importantly, we show that these possible failure modes are not due to the lack of expressivity in the NN architecture, but that the PINN's setup makes the loss landscape very hard to optimize. We then describe two promising solutions to address these failure modes. The first approach is to use curriculum regularization, where the PINN's loss term starts from a simple PDE regularization, and becomes progressively more complex as the NN gets trained. The second approach is to pose the problem as a sequence-to-sequence learning task, rather than learning to predict the entire space-time at once. Extensive testing shows that we can achieve up to 1-2 orders of magnitude lower error with these methods as compared to regular PINN training.

22 pages

References in corpus (7)

• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• Understanding and mitigating gradient pathologies in physics-informed neural networks
• When and why PINNs fail to train: A neural tangent kernel perspective
• DC3: A learning method for optimization with hard constraints
• Multi-scale Deep Neural Network (MscaleDNN) Methods for Oscillatory Stokes Flows in Complex Domains
• NVIDIA SimNet^{TM}: an AI-accelerated multi-physics simulation framework
• Physics Constrained Learning for Data-driven Inverse Modeling from Sparse Observations

Cited by in corpus (8)

• Physical Activation Functions (PAFs): An Approach for More Efficient Induction of Physics into Physics-Informed Neural Networks (PINNs)
• Differential Walk on Spheres
• Adapting Physics-Informed Neural Networks to Improve ODE Optimization in Mosquito Population Dynamics
• Multi-Fidelity Machine Learning Applied to Steady Fluid Flows
• Fast PDE-constrained optimization via self-supervised operator learning
• Neural Hawkes: Non-Parametric Estimation in High Dimension and Causality Analysis in Cryptocurrency Markets
• Neuroevolution of Physics-Informed Neural Nets: Benchmark Problems and Comparative Results
• Learning from learning machines: a new generation of AI technology to meet the needs of science