arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.05652cs.LGcs.AI
收藏

“这些数据集之间有何不同?”一个解释数据分布偏移的框架

"What is Different Between These Datasets?" A Framework for Explaining Data Distribution Shifts

Department of Computer Science Duke University(计算机科学系 哥伦比亚大学);Department of Electrical and Computer Engineering Duke University(电气与计算机工程系 哥伦比亚大学)

Varun Babbar, Zhicheng Guo, Cynthia Rudin

AI总结:

本文提出一个可解释的通用框架,用于比较数据集并解释分布偏移,在表格、文本、图像和时间序列等多种模态的低维与高维场景中验证了其有效性,为理解和应对分布偏移提供了可操作的洞察。

摘要:

机器学习模型的性能在很大程度上依赖于输入数据的质量,然而现实世界中的应用常常面临与数据相关的重大挑战。一个常见问题出现在整理训练数据或部署模型时:来自同一领域的两个数据集可能呈现不同的分布。尽管已有许多技术用于检测此类分布偏移,但除了不透明的定量指标之外,仍缺乏以人类可理解的方式解释这些差异的综合性方法。为了弥合这一差距,我们提出了一个通用的可解释方法框架,用于比较数据集。通过各种案例研究,我们证明了该方法在多种数据模态上的有效性——包括表格数据、文本数据、图像、时间序列信号——涵盖低维和高维设置。这些方法通过提供可操作且可解释的洞察来补充现有技术,从而更好地理解和应对分布偏移。

英文摘要:

The performance of machine learning models relies heavily on the quality of input data, yet real-world applications often face significant data-related challenges. A common issue arises when curating training data or deploying models: two datasets from the same domain may exhibit differing distributions. While many techniques exist for detecting such distribution shifts, there is a lack of comprehensive methods to explain these differences in a human-understandable way beyond opaque quantitative metrics. To bridge this gap, we propose a versatile framework of interpretable methods for comparing datasets. Using a variety of case studies, we demonstrate the effectiveness of our approach across diverse data modalities-including tabular data, text data, images, time-series signals -- in both low and high-dimensional settings. These methods complement existing techniques by providing actionable and interpretable insights to better understand and address distribution shifts.

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。

↑