{"id":"arxiv/2610.03442","url":"https://arxiv.org/abs/2610.03442","tags":["math.NT"],"score":0,"title":"Hausdorff dimension for exactly approximable matrices","domain":"math.NT","source":"arxiv","authors":["Yubin He","Bixuan Li","Baowei Wang"],"reviews":0,"subject":"math.NT","summary":"We establish the Hausdorff dimension for exactly approximable matrices for all dimensions, including the inhomogeneous case. More precisely, let $ψ:\\N\\to(0,\\infty)$ be a non-increasing approximation function and let $τ_ψ$ denote its lower order. For $m,n\\ge1$ and $\\boldsymbolθ\\in\\mathbb R^m$, denote $\\Exact_{m,n}(ψ,\\boldsymbolθ)$ the set of $m\\times n$ matrices that are $ψ$-approximable with shift $\\boldsymbolθ$, but are not $cψ$-approximable for any $0 \\frac{n}{m}. \\] We adopt a probabilistic viewpoint, i.e. by an averaged counting argument to quantify the distribution of rational vectors…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-02","reactions":{}}