Least-Squares ReLU Neural Network (LSNN) Method For Linear Advection-Reaction Equation
This is an incremental improvement for computational physics and numerical analysis, offering a more efficient neural network-based solver for discontinuous PDEs.
The paper tackles solving linear advection-reaction equations with discontinuous solutions by proposing a least-squares ReLU neural network method, which automatically approximates discontinuities and outperforms mesh-based methods in parameter efficiency while avoiding Gibbs phenomena.
This paper studies least-squares ReLU neural network method for solving the linear advection-reaction problem with discontinuous solution. The method is a discretization of an equivalent least-squares formulation in the set of neural network functions with the ReLU activation function. The method is capable of approximating the discontinuous interface of the underlying problem automatically through the free hyper-planes of the ReLU neural network and, hence, outperforms mesh-based numerical methods in terms of the number of degrees of freedom. Numerical results of some benchmark test problems show that the method can not only approximate the solution with the least number of parameters, but also avoid the common Gibbs phenomena along the discontinuous interface. Moreover, a three-layer ReLU neural network is necessary and sufficient in order to well approximate a discontinuous solution with an interface in $\mathbb{R}^2$ that is not a straight line.