arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.02629cs.LG
收藏

Effector:一个用于区域解释的Python包

Effector: A Python Package for Global and Regional Feature Effects

Harokopio University of Athens(哈尔科波伊欧大学);ATHENA Research Center(ATHENA研究中心);Leibniz Institute for Prevention Research and Epidemiology - BIPS(莱布尼茨预防研究与流行病学研究所 - BIPS);University of Warsaw(华沙大学);University of the Bundeswehr Munich(联邦国防军大学慕尼黑);LMU Munich and Munich Center for Machine Learning (MCML)(慕尼黑大学和慕尼黑机器学习中心 (MCML))
Vasilis Gkolemis, Christos Diou, Dimitris Kyriakopoulos, Konstantinos Tsopelas, Julia Herbinger, Hubert Baniecki, Dimitrios Rontogiannis, Loukas Kavouras, Maxim… 展开作者

Vasilis Gkolemis, Christos Diou, Dimitris Kyriakopoulos, Konstantinos Tsopelas, Julia Herbinger, Hubert Baniecki, Dimitrios Rontogiannis, Loukas Kavouras, Maximilian Muschalik, Theodore Dalamagas, Eirini Ntoutsi, Bernd Bischl, Giuseppe Casalicchio

AI总结:

针对表格机器学习模型全局特征效应在特征交互时易误导的问题,开发开源Python包Effector,提供统一API下高效的全局与区域特征效应实现,兼容主流ML库且易扩展。

摘要:

Effector是一个Python包,用于通过全局和区域特征效应解释基于表格数据训练的机器学习(ML)模型。全局效应方法如部分依赖图(PDP)和累积局部效应(ALE)因形式简洁被广泛用于解释表格ML模型——每个特征对预测结果的平均影响可通过单张一维图呈现。但当特征间存在交互作用时,全局效应可能产生误导。区域效应方法通过将输入空间划分为内部交互作用极小的不相交子区域,为每个子空间计算独立的区域效应来解决该问题,最终每个特征的区域效应通过一组一维图可视化。Effector在统一API下提供了当前最优的全局与区域特征效应方法的高效实现,可与scikit-learn、PyTorch等主流ML库无缝集成,具备模块化、可扩展的设计,附带完善的文档与教程,是开源项目,已在Github公开,地址为https://github.com/givasile/effector。

英文摘要:

Effector is a Python package for interpreting machine learning (ML) models that are trained on tabular data through global and regional feature effects. Global effects, like Partial Dependence Plot (PDP) and Accumulated Local Effects (ALE), are widely used for explaining tabular ML models due to their simplicity -- each feature's average influence on the prediction is summarized by a single 1D plot. However, when features are interacting, global effects can be misleading. Regional effects address this by partitioning the input space into disjoint subregions with minimal interactions within each and computing a separate regional effect per subspace. Regional effects are then visualized by a set of 1D plots per feature. Effector provides efficient implementations of state-of-the-art global and regional feature effects methods under a unified API. The package integrates seamlessly with major ML libraries like scikit-learn and PyTorch. It is designed to be modular and extensible, and comes with comprehensive documentation and tutorials. Effector is an open-source project publicly available on Github at https://github.com/givasile/effector.

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。

↑