On the Compatibility between Neural Networks and Partial Differential Equations for Physics-informed Learning
This addresses a fundamental limitation in physics-informed machine learning for solving PDEs, offering a novel architecture that enforces point-wise correctness, though it is incremental in improving PINN methods.
The paper identifies that ReLU-based MLPs cannot solve second- or higher-order PDEs due to a vanished Hessian, and proposes a physics-enforced PINN architecture using C^n activation functions and an output-layer hyperplane to strictly satisfy linear PDEs without a PDE loss term.
We shed light on a pitfall and an opportunity in physics-informed neural networks (PINNs). We prove that a multilayer perceptron (MLP) only with ReLU (Rectified Linear Unit) or ReLU-like Lipschitz activation functions will always lead to a vanished Hessian. Such a network-imposed constraint contradicts any second- or higher-order partial differential equations (PDEs). Therefore, a ReLU-based MLP cannot form a permissible function space for the approximation of their solutions. Inspired by this pitfall, we prove that a linear PDE up to the $n$-th order can be strictly satisfied by an MLP with $C^n$ activation functions when the weights of its output layer lie on a certain hyperplane, as called the out-layer-hyperplane. An MLP equipped with the out-layer-hyperplane becomes "physics-enforced", no longer requiring a loss function for the PDE itself (but only those for the initial and boundary conditions). Such a hyperplane exists not only for MLPs but for any network architecture tailed by a fully-connected hidden layer. To our knowledge, this should be the first PINN architecture that enforces point-wise correctness of PDEs. We show a closed-form expression of the out-layer-hyperplane for second-order linear PDEs, which can be generalised to higher-order nonlinear PDEs.