提出可衡量数据解释效率的新指标，评估数据对模型可解释性的贡献。
Interpretive Efficiency: Information-Geometric Foundations of Data Usefulness
- 基于信息几何构建任务感知的解释效率度量，满足五条合理性公理。
- 在图像与信号任务中验证其能识别冗余表示并关联模型鲁棒性。
- 适合关注模型可解释性与表征设计的研究者使用。
可解释性是可信机器学习的核心，但现有度量很少量化数据支持解释性表征的有效性。本文提出解释效率（Interpretive Efficiency），一个归一化、任务感知的函数，用于衡量通过解释通道传递的任务相关信息比例。该定义基于五个公理：有界性、黑威尔式单调性、数据处理稳定性、可接受不变性和渐近一致性。我们将其与互信息关联，并推导出局部费舍尔几何展开，利用标准经验过程工具建立了渐近和有限样本估计保证。在受控的图像与信号任务实验中表明，该度量能恢复理论排序，揭示被准确率掩盖的表示冗余，并与鲁棒性相关，是一种兼具理论基础与实用价值的表征设计诊断工具。
原文摘要 · Abstract (English)
Interpretability is central to trustworthy machine learning, yet existing metrics rarely quantify how effectively data support an interpretive representation. We propose Interpretive Efficiency, a normalized, task-aware functional that measures the fraction of task-relevant information transmitted through an interpretive channel. The definition is grounded in five axioms ensuring boundedness, Blackwell-style monotonicity, data-processing stability, admissible invariance, and asymptotic consistency. We relate the functional to mutual information and derive a local Fisher-geometric expansion, then establish asymptotic and finite-sample estimation guarantees using standard empirical-process tools. Experiments on controlled image and signal tasks demonstrate that the measure recovers theoretical orderings, exposes representational redundancy masked by accuracy, and correlates with robustness, making it a practical, theory-backed diagnostic for representation design.
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。