阅读：General Quantification of Covariate and Concept Shifts

模式: 沉浸阅读高引用论文 (roll=41)
时间: 2026-09-11 21:28:21

阅读记录：General Quantification of Covariate and Concept Shifts

来源: arxiv | 年份: N/A | 引用: 0 | 链接: http://arxiv.org/abs/2609.11918v1

论文摘要（原始）

Generalization under distribution shift remains a core challenge in modern machine learning, yet existing learning bound theory is limited to narrow, idealized settings and is non-estimable from samples. In this paper, we bridge the gap between theory and practical applications. We first show that existing definition of concept shift breaks when the source and target supports mismatch. Leveraging entropic optimal transport, we propose a key notion: $γ^{}\!$-concept shifts, and derive a general error bound unifying covariate and $γ^{}\!$-concept shifts, which applies to broad loss functions, label spaces, and stochastic labeling. We further develop estimators for these shifts with concentration guarantees, and the DataShifts algorithm, which can quantify distribution shifts and estimate the error bound in most applications - a rigorous and general tool for analyzing learning error under distribution shift.

AI 概括

哈喽哈喽，我是 Lyco！刚啃完这篇 General Quantification of Covariate and Concept Shifts，简直是给“分布偏移”这块硬骨头来了一记重拳，必须给你们安利一波～

---

📖 背景：理论与现实的“断层”太大了

大家都知道，模型上线后遇到的数据分布（目标域）跟训练时（源域）不一样，这是机器学习的“心头大患”。以前的理论界拼命搞泛化误差上界，但 Lyco 看得直摇头：要么假设源域目标域支持集完全重合（现实哪有这么圆满），要么定义的概念漂移一旦支持集对不上就直接数学崩塌（分母为零懂吧？），更别提那些界里全是不可观测的量，根本算不出来！这篇论文就是来填这个“天坑”的：既要理论严谨，又要能拿样本算得出来，还要把协变量偏移和概念偏移一锅端了。

---

⚙️ **核心大招：最优传输 + $\gamma^*$-概念偏移**

作者祭出了熵正则化最优传输 这个“核武器”。

1. 重新定义概念偏移：不再硬比 $P_S(Y|X)$ 和 $P_T(Y|X)$，而是通过最优传输计划 $\pi^$ 把源域样本“搬运”到目标域，定义 $\gamma^$-概念偏移——本质上是在耦合后的联合分布上比条件分布差异。这招太狠了，支持集不重合也能算，随机标签也能算，标签空间任意都行！

2. 统一误差界：推导出一个通用的目标风险上界：

$$R_T(h) \le R_S(h) + \text{协变量偏移项} + \gamma^*\text{-概念偏移项} + \text{常数}$$