Id: arxiv/2607.25031 Url: https://arxiv.org/abs/2607.25031 Tags: math.ST, stat.ME, stat.ML Score: 0 Title: Transfer Learning in High-Dimensional Clustering: Minimax Thresholds and Applications in Single-Cell Data Domain: math.ST Source: arxiv Authors: Abhinav Chakraborty, Sagnik Nandy Reviews: 0 Subject: math.ST Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-07-27 Page: https://delphi.science/r/arxiv/2607.25031 Clustering is a fundamental problem in statistics, with applications across many scientific disciplines. In many modern applications involving clustering, the primary dataset (the target data) is accompanied by related datasets (the source data). Transferring information from such sources may improve clustering accuracy in the target, making transfer learning for clustering practically important. Despite recent progress, the conditions under which source data improve target clustering remain unclear in high-dimensional settings, even for the canonical Gaussian mixture model. In this paper,…