arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.03791cs.LGcs.CY
收藏

是否应该用合成数据?对合成数据用于数据共享和增强的适用性分析

Should I use Synthetic Data for That? An Analysis of the Suitability of Synthetic Data for Data Sharing and Augmentation

Lausanne University Hospital(拉索纳大学医院);Swiss Data Science Center, EPFL(瑞士数据科学中心,EPFL);University of Lausanne(洛桑大学);Max Planck Institute for Security & Privacy EPFL(马克斯·普朗克安全与隐私研究所 EPFL)

Bogdan Kulynych, Theresa Stadler, Jean Louis Raisaro, Carmela Troncoso

AI总结:

本文分析了合成数据在数据共享和增强中的适用性,指出其在实际应用中存在基本和实际限制,导致许多现有或设想的使用案例并不适合。

摘要:

近年来,生成建模的进步使许多人将合成数据视为解决数据访问、稀缺性和代表性不足问题的首选解决方案。在本文中,我们研究了三个突出的使用案例:(1)将合成数据作为专有数据集的代理,以在保护隐私的同时进行统计分析;(2)通过合成数据增强机器学习训练集以提高模型性能;(3)通过合成数据增强数据集以减少统计估计中的方差。对于每个使用案例,我们正式定义了问题设置,并通过形式分析和案例研究,研究在何种条件下合成数据可以实现其预期目标。我们识别出限制合成数据在特定问题中有效作为解决方案的基本和实际限制。我们的分析揭示,由于这些限制,许多现有或设想的合成数据使用案例都存在较差的问题适配性。我们的形式化和对合成数据使用案例的分类使决策者能够评估合成数据是否适合他们特定的数据可用性问题。

英文摘要:

Recent advances in generative modelling have led many to see synthetic data as the go-to solution for a range of problems around data access, scarcity, and under-representation. In this paper, we study three prominent use cases: (1) Sharing synthetic data as a proxy for proprietary datasets to enable statistical analyses while protecting privacy, (2) Augmenting machine learning training sets with synthetic data to improve model performance, and (3) Augmenting datasets with synthetic data to reduce variance in statistical estimation. For each use case, we formalise the problem setting and study, through formal analysis and case studies, under which conditions synthetic data can achieve its intended objectives. We identify fundamental and practical limits that constrain when synthetic data can serve as an effective solution for a particular problem. Our analysis reveals that due to these limits many existing or envisioned use cases of synthetic data are a poor problem fit. Our formalisations and classification of synthetic data use cases enable decision makers to assess whether synthetic data is a suitable approach for their specific data availability problem.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑