{"id":"arxiv/2603.14383","url":"https://arxiv.org/abs/2603.14383","tags":["eess.SP","math.DS"],"score":0,"title":"Geometric Mode-Selection Scores for Delay-Coordinates Dynamic Mode Decomposition","domain":"eess.SP","source":"arxiv","authors":["Yoav Harris","Hadas Benisty","Ronen Talmon"],"reviews":0,"subject":"eess.SP","summary":"Delay-coordinates dynamic mode decomposition (DC-DMD) is widely used to extract coherent spatiotemporal modes from high-dimensional time series. A central challenge is distinguishing dynamically meaningful modes from spurious modes induced by noise and order overestimation. We frame this as a mode-selection scoring problem: each mode receives a score that ranks it as true or spurious; any hard selection (threshold or clustering) is a downstream choice. We show that mode selection in DC-DMD is fundamentally a problem of subspace geometry. True modes are characterized by concentration within a…","category":"research","comments":0,"disputed":false,"division":"eess","mentions":0,"discussed":0,"published":"2026-03-15","reactions":{}}