提出新方法提升变量重要性推断的稳定性与准确性
Targeted Learning for Variable Importance
- 基于目标学习框架，改进条件置换重要性度量的推断
- 在有限样本下显著提升精度，同时保持高效计算
- 适合关注模型可解释性与统计稳健性的研究者
变量重要性是机器学习可解释性中最广泛使用的度量之一，受到统计学和机器学习领域的广泛关注。近年来，对这些度量的不确定性量化日益受到重视。现有方法大多依赖单步程序，虽在渐近意义上高效，但在有限样本下敏感性高、不稳定。为此，我们提出一种基于目标学习（Targeted Learning, TL）框架的新方法，旨在增强变量重要性度量推断的鲁棒性。该方法特别适用于条件置换变量重要性。我们证明其 (i) 保持传统方法的渐近效率，(ii) 计算复杂度相当，(iii) 在有限样本条件下显著提升准确性。数值实验进一步验证了方法的实际优势，并支持理论结果。
原文摘要 · Abstract (English)
Variable importance is one of the most widely used measures for interpreting machine learning with significant interest from both statistics and machine learning communities. Recently, increasing attention has been directed toward uncertainty quantification in these metrics. Current approaches largely rely on one-step procedures, which, while asymptotically efficient, can present higher sensitivity and instability in finite sample settings. To address these limitations, we propose a novel method by employing the targeted learning (TL) framework, designed to enhance robustness in inference for variable importance metrics. Our approach is particularly suited for conditional permutation variable importance. We show that it (i) retains the asymptotic efficiency of traditional methods, (ii) maintains comparable computational complexity, and (iii) delivers improved accuracy, especially in finite sample contexts. We further support these findings with numerical experiments that illustrate the practical advantages of our method and validate the theoretical results.
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。