{"id":"arxiv/2605.31583","url":"https://arxiv.org/abs/2605.31583","tags":["math.ST","math.PR"],"score":0,"title":"Sharp minimax risks and phase transitions in sparse submatrix detection","domain":"math.ST","source":"arxiv","authors":["Subhajit Goswami","Rajarshi Mukherjee"],"reviews":0,"subject":"math.ST","summary":"We study the minimax risk for detecting a sparse elevated-mean Gaussian submatrix inside a larger noisy matrix. When the planted submatrix has size $n\\times n$ and the ambient matrix has size $N\\times N$ with $N = n^{1+α}$, the classical work of \\cite{butuceasubmatrix2013} identifies the sharp detection boundary around which the minimax risk converges to $0$ or $1$. This paper extends that zero-one theory by determining the precise asymptotic rate of the minimax risk throughout a two-variable phase diagram. Above the detection boundary, we determine the precise exponent for the stretched or…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-05-29","reactions":{}}