{"id":"arxiv/2607.28981","url":"https://arxiv.org/abs/2607.28981","tags":["stat.ME","econ.EM","math.ST","stat.AP","stat.ML"],"score":0,"title":"Distance Profile Embedding for Independence and Conditional Independence Testing of Random Objects","domain":"stat.ME","source":"arxiv","authors":["Wenxi Tan","Bing Li","Lingzhou Xue"],"reviews":0,"subject":"stat.ME","summary":"Testing independence or conditional independence is fundamental to statistical inference, yet existing methods for non-Euclidean random objects often face a difficult trade-off between geometric flexibility and theoretical tractability. We introduce the Distance Profile Embedding (DPE), a novel representation that maps random objects from general metric spaces into a Hilbert space of square-integrable functions. We prove that this mapping is injective and preserves full distributional information without requiring isometric Hilbert embeddings or one-to-one correspondence conditions.…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-07-31","reactions":{}}