{"id":"arxiv/2607.17664","url":"https://arxiv.org/abs/2607.17664","tags":["eess.SY","math.DS","q-bio.QM"],"score":0,"title":"Graph-Induced Tensor Liftings for Networked SEIR Models: Dimensional Reduction and Residual Analysis","domain":"eess.SY","source":"arxiv","authors":["Enrique Baeyens"],"reviews":0,"subject":"eess.SY","summary":"Networked SEIR models describe epidemic spread within and between interacting subpopulations through contact-supported nonlinear transmission. Standard polynomial liftings based on complete ordered Kronecker tensors yield linear higher-dimensional representations, but their dimensions grow rapidly because they retain interactions absent from the transmission graph. This paper develops a graph-induced tensor lifting whose observables are selected from the effective transmission support. An exact edge-based quadratic representation separates linear compartmental transitions from nonlinear…","category":"research","comments":0,"disputed":false,"division":"eess","mentions":0,"discussed":0,"published":"2026-07-20","reactions":{}}