{"id":"arxiv/2609.04822","url":"https://arxiv.org/abs/2609.04822","tags":["stat.ML","cs.LG","math.ST"],"score":0,"title":"Minimax Lower Bound for Estimating Diffusion-based Local Intrinsic Dimension","domain":"stat.ML","source":"arxiv","authors":["Jaehee Seo","Wontae Jeong","Jisu Kim"],"reviews":0,"subject":"stat.ML","summary":"While diffusion-based methods have recently emerged as effective tools for probing the intrinsic geometry of high-dimensional data, their statistical difficulty remains largely unexplored. We study estimation of the finite-scale population functional underlying FLIPD (Kamkari et al., 2024; arXiv:2406.03537), a diffusion-based local intrinsic dimension (LID) quantity defined through the logarithmic scale derivative of a Gaussian-smoothed density. Intuitively, Gaussian smoothing turns local dimension into a scale law: near a $d$-dimensional manifold, the kernel mass grows like $σ^d$, so…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-09-04","reactions":{}}