用条件距离相关性抑制深度学习中的数据偏见，提升模型鲁棒性。
DISCO: Mitigating Bias in Deep Learning with Conditional Distance Correlation
- 基于因果框架构建条件独立性准则，识别并消除偏见机制。
- 在六个数据集上优于或媲美现有方法，且超参数更少。
- 适合关注模型公平性与稳健性的研究人员和工程师。
数据偏见常导致深度学习模型依赖虚假相关而非任务相关信号。本文提出标准反因果模型（SAM），一个统一的因果框架，用于刻画偏见机制并推导因果稳定性的条件独立性准则。基于该理论，我们设计了DISCO$_m$和sDISCO，高效可扩展的条件距离相关估计器，支持梯度模型中的独立性正则化。在六个不同数据集上，所提方法持续优于或媲美现有观测偏见缓解方法，同时所需超参数更少，并能无缝扩展至多偏见场景。本工作连接因果理论与实际深度学习，为鲁棒预测提供理论基础与有效工具。源代码：https://github.com/yakamoz5/DISCO。
原文摘要 · Abstract (English)
Dataset bias often leads deep learning models to exploit spurious correlations instead of task-relevant signals. We introduce the Standard Anti-Causal Model (SAM), a unifying causal framework that characterizes bias mechanisms and yields a conditional independence criterion for causal stability. Building on this theory, we propose DISCO$_m$ and sDISCO, efficient and scalable estimators of conditional distance correlation that enable independence regularization in gradient-based models. Across six diverse datasets, our methods consistently outperform or are competitive in existing observed bias mitigation approaches, while requiring fewer hyperparameters and scaling seamlessly to multi-bias scenarios. This work bridges causal theory and practical deep learning, providing both a principled foundation and effective tools for robust prediction. Source Code: https://github.com/yakamoz5/DISCO.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。