{"id":"arxiv/2609.30477","url":"https://arxiv.org/abs/2609.30477","tags":["math.ST","cs.LG"],"score":0,"title":"Scaffold-Constrained Subset Dynamic Programming for Exact SSE Clustering","domain":"math.ST","source":"arxiv","authors":["Yordan P. Raykov","Max A. Little"],"reviews":0,"subject":"math.ST","summary":"Exact Euclidean \\(K\\)-means partitions \\(n\\) observations into \\(K\\) unlabelled clusters, but the unrestricted search is generally exponential. We use data-derived geometric graphs to precondition an exact subset dynamic program: as a result only connected vertex subsets are admitted as clusters, while sum-of-squared-errors (SSE) loss is unchanged. A remaining-set recurrence minimises fixed-\\(K\\) or penalised SSE, with exact factorisation over the connected components of each remaining set. The central question we study is how much computational support can be removed while preserving an…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-09-24","reactions":{}}