愿论文：Neural Operator: Graph Kernel Network for Partial Differential Equations. 原作者英文blog：Graph Neural Operator for PDEs. 引言. 科学计算 的成本非常高。数值求解器模拟流体动力学和 多体运动 可能需要几天甚至几个月。

DIFFERENTIAL EQUATIONS ON GRAPHS OLIVER KNILL, HCRP PROJECT WITH ANNIE RAK Abstract. We look at examples of dynamical systems on nite simple graphs. These systems correspond to partial di erential equations in the continuum. This documents contains some notes related to a project supported by the HCRP Summer 2016 program

In general, we will consider a partial differential equation in a scalar function u(x) to be any equation of the form F(fDnu : n 2Z+g, x) = G(x), where Dnu denotes the set of nth derivatives of u. To simplify matters, we will consider only the case where F is linear in u and its derivatives, thus denoting a linear PDE.

Partial differential equations on graphs This project with Annie Rak started in the summer 2016 as a HCRP project. The topic is ``differential equations on graphs". We explored in the summer 2016 first various dynamical systems on networks.

Multipole Graph Neural Operator for Parametric Partial Differential Equations. NeurIPS20. Inspired by the fast multipole method (FMM), we propose a novel hierarchical, and multi-scale graph structure which, when deployed with GNNs, captures global properties of the PDE solution operator with a linear time-complexity.