跳到正文
arXiv cs.AI· Mohammad Tariqul Islam, Du Liu, Deblina Sarkar·· 4 小时前AI 评分38

Manifold Approximation 带来更稳健的核对齐:MKA 度量提出

Manifold Approximation leads to Robust Kernel Alignment

AI 导读

研究者提出 Manifold-approximated Kernel Alignment(MKA),将局部邻域等流形几何信息引入表示对齐度量,以弥补 CKA 不考虑底层流形且依赖大量启发式设置的缺陷。

正文 · 原文

View PDF HTML (experimental)

Abstract:Centered kernel alignment (CKA) is a popular metric for comparing representations, determining equivalence of networks, and conducting neuroscience research. However, CKA does not account for the underlying manifold and relies on numerous heuristics that cause it to behave differently at different scales of data. In this work, we propose Manifold-approximated Kernel Alignment (MKA), which incorporates manifold geometry (i.e., local neighborhood) into the alignment task. We derive a theoretical framework for MKA. Our construction removes representation-dependent degree weighting, yields a closed-form alignment score, and gives each row a probabilistic interpretation as a local Markov transition distribution. We further establish kernel- and score-level stability with respect to the neighborhood size. We perform empirical evaluations on synthetic datasets and real-world examples to characterize and compare MKA to its contemporaries. In controlled synthetic experiments, MKA is consistently less sensitive to hyperparameters and captures the selected topology-preserving relationships. On the large-scale benchmark, MKA achieves its strongest aggregate performance in vision, is competitive on graphs, and is weaker than CKA in several NLP evaluations. Our findings suggest that manifold-aware kernel alignment provides a robust foundation for measuring representations and is potentially useful when local neighborhood geometry is important, with possible applications in representation learning.
Comments: 9 pages, 5 figures + supplementary
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
Cite as: arXiv:2510.22953 [cs.LG]
  (or arXiv:2510.22953v3 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.2510.22953

arXiv-issued DOI via DataCite

Journal reference: Transactions on Machine Learning Research, 2026

Submission history

From: Mohammad Tariqul Islam [view email]
[v1] Mon, 27 Oct 2025 03:16:15 UTC (7,191 KB)
[v2] Sun, 18 Oct 2026 20:06:57 UTC (8,410 KB)
[v3] Fri, 13 Nov 2026 15:29:34 UTC (8,419 KB)

来源:arXiv cs.AI · arxiv.org