arXiv：cs.LG· Nir Ben-Ari, Ronen Talmon, Uri Shaham·· 4 小时前AI 评分35
Unpaired CCA（UCCA）：无需配对数据的典型相关分析方法
Unpaired Canonical Correlation Analysis
AI 导读
研究者提出 Unpaired CCA（UCCA），一种在训练中完全不使用配对样本、仅凭非配对数据学习线性投影以最大化真实潜在配对相关性的方法，被接收至 NeurIPS 2026。该工作首次建立二次分配问题（QAP）与 CCA 的理论联系，并据此推导出实用算法，是严格非配对设定下学习最大相关投影的首个方法。在真实多模态数据集上，UCCA 在恢复潜在真实相关性方面显著优于近期的非配对对齐基线。
正文
Abstract:Canonical Correlation Analysis (CCA) is a fundamental method for multiview shared space learning. However, its strict reliance on paired data poses a significant limitation, as such data is often difficult to obtain or entirely unavailable. In this paper, we present Unpaired CCA (UCCA), a novel method that learns linear projections to maximize the correlation of the true underlying pairing without access to any paired samples during training. We first establish theoretical results connecting the Quadratic Assignment Problem (QAP) to CCA. Leveraging these theoretical insights, we derive a practical method to maximize correlation exclusively from unpaired data. To the best of our knowledge, UCCA is the first approach to learn maximally correlated projections in a strictly unpaired setting. We validate UCCA on real-world multi-modal datasets, demonstrating that it significantly outperforms recent unpaired alignment baselines in recovering the underlying true correlation. This work fills a critical gap between traditional statistical multiview learning and the growing field of unpaired data learning.
| Comments: | Accepted to NeurIPS 2026 |
| Subjects: | Machine Learning (stat.ML); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.09530 [stat.ML] |
| (or arXiv:2610.09530v1 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2610.09530 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Nir Ben-Ari [view email]
[v1]
Wed, 7 Oct 2026 06:30:15 UTC (2,884 KB)
来源：arXiv：cs.LG · arxiv.org