Authors
Danielle C Maddix, Nadim Saad, Yuyang Wang
Publication date
2021/12
Conference
Machine Learning and the Physical Sciences (MLPS) NeurIPS 2021 Workshop
Description
The transport of traffic flow can be modeled by the advection equation. Finite difference and finite volumes methods have been used to numerically solve this hyperbolic equation on a mesh. Advection has also been modeled discretely on directed graphs using the graph advection operator [4, 18]. In this paper, we first show that we can reformulate this graph advection operator as a finite difference scheme. We then propose the Directed Graph Advection Mat\'ern Gaussian Process (DGAMGP) model that incorporates the dynamics of this graph advection operator into the kernel of a trainable Mat\'ern Gaussian Process to effectively model traffic flow and its uncertainty as an advective process on a directed graph.
Total citations
20232
Scholar articles
DC Maddix, N Saad, Y Wang - arXiv preprint arXiv:2201.00001, 2021