arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2412.05152cs.LGcs.AI
收藏

应对捷径、虚假关联与混淆因素:从起源、检测到缓解

Navigating Shortcuts, Spurious Correlations, and Confounders: From Origins via Detection to Mitigation

TU Darmstadt(达姆施塔特工业大学);Hessian Center for AI (hessian.AI)(黑森人工智能中心);German Research Center for AI (DFKI)(德国人工智能研究中心);Centre for Cognitive Science, TU Darmstadt(达姆施塔特工业大学认知科学中心)

David Steinmann, Felix Divo, Maurice Kraus, Antonia Wüst, Lukas Struppek, Felix Friedrich, Kristian Kersting

AI总结:

本文提出捷径学习的统一定义与分类体系,贯通多元术语及偏见、因果和安全领域,并梳理检测、缓解方法与专用数据集。

摘要:

捷径(shortcuts)也被描述为 Clever Hans 行为、虚假关联或混淆因素,是机器学习与人工智能中的一项重大挑战,会严重影响模型的泛化能力和稳健性。然而,这一领域的研究仍分散在各种不同术语之下,阻碍了整个领域的进展。因此,我们通过给出捷径的形式化定义并打通文献中使用的多元术语,提出了一个关于捷径学习的统一分类体系。在此过程中,我们进一步建立了捷径与相关领域之间的重要联系,包括偏见、因果性和安全性;这些领域之间存在相似之处,却很少被讨论。我们的分类体系梳理了现有的捷径检测与缓解方法,全面概述了该领域的当前状态,并揭示了研究不足的方向和开放性挑战。此外,我们还汇编并分类了专为研究捷径学习而设计的数据集。总体而言,本工作提供了一个整体性视角,以加深理解并推动开发更有效的策略,从而应对机器学习中的捷径问题。

英文摘要:

Shortcuts, also described as Clever Hans behavior, spurious correlations, or confounders, present a significant challenge in machine learning and AI, critically affecting model generalization and robustness. Research in this area, however, remains fragmented across various terminologies, hindering the progress of the field as a whole. Consequently, we introduce a unifying taxonomy of shortcut learning by providing a formal definition of shortcuts and bridging the diverse terms used in the literature. In doing so, we further establish important connections between shortcuts and related fields, including bias, causality, and security, where parallels exist but are rarely discussed. Our taxonomy organizes existing approaches for shortcut detection and mitigation, providing a comprehensive overview of the current state of the field and revealing underexplored areas and open challenges. Moreover, we compile and classify datasets tailored to study shortcut learning. Altogether, this work provides a holistic perspective to deepen understanding and drive the development of more effective strategies for addressing shortcuts in machine learning.

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。

↑