We identify first-mover bias -- path-dependent concentration of SHAP feature importance from sequential residual fitting in gradient boosting -- as a mechanistic contributor to attribution instability under multicollinearity. Scaling up a single model amplifies this effect: a Large Single Model matching our method's total tree count produces the poorest attribution reproducibility of any approach tested.
We show that model independence largely neutralizes first-mover bias. Both DASH (Diversified Aggregation of SHAP) and simple seed-averaging (Stochastic Retrain) restore stability by breaking the sequential dependency chain. At rho=0.9, both achieve stability ~0.977, while Single Best degrades to 0.958 and LSM to 0.938. On Breast Cancer, DASH improves stability from 0.376 to 0.925 (+0.549), outperforming Stochastic Retrain by +0.063. Under nonlinear DGPs, the advantage emerges at rho>=0.7.
DASH provides two diagnostic tools -- the Feature Stability Index and Importance-Stability Plot -- that detect first-mover bias without ground truth. A crossed ANOVA with formal F-statistics confirms the mechanism: DASH shifts variance from model-dominated (40.6%) to data-dominated (73.6%). Software at https://github.com/DrakeCaraker/dash-shap
Margins are Insufficient for Explaining Gradient Boosting
Aarhus University(奥胡斯大学)
Boosting是机器学习中最成功的想法之一,几乎不需要微调就可以实现很好的实际性能。提升分类器的成功通常归因于边缘的改善。对边缘解释的关注是由Schapire et al.(1998)开创的开创性工作,并在Gao和Zhou(2013)的$k$的边缘概括中达到高潮,最近被证明对于某些数据分布来说是接近紧密的(Gronlund et al. 2019)。在这项工作中,我们首先证明了$k$的第三边界是不够的,在解释国家的最先进的梯度助推器的性能。然后,我们解释了短期的$k$ 'th边缘界,并证明了一个更强大,更精细的边缘为基础的推广界的提升分类器,确实成功地解释了现代梯度助推器的性能。最后,我们改进了Grønlund et al.(2019)最近的泛化下界。
Explainable Incipient Fault Detection Systems for Photovoltaic Panels
Kalasalingam Academy for Research and Education(卡萨斯林甘研究与教育学院);GE Corporate Research Center(通用电气企业研究中心);SINTEF;Norwegian University of Science and Technology(挪威科技大学);SUNY Polytechnic Institute(纽约州立大学理工学院)
Financial Fraud Detection Using Explainable AI and Stacking Ensemble Methods
Computer Science Department, College of Computers and Information Technology, Taif University(计算机科学系,计算机与信息科技学院,泰夫大学);Department of Computer Science, Taif University(计算机科学系,泰夫大学)
McGill University, Department of Mathematics and Statistics(麦吉尔大学数学与统计学系);Université Marie et Louis Pasteur, CNRS, LmB (UMR 6623)(玛丽·路易斯·帕斯特大学、法国国家科学研究中心、LmB(UMR 6623))
Inductive inference of gradient-boosted decision trees on graphs for insurance fraud detection
Data Office, Allianz Benelux(Allianz Benelux 数据办公室);KU Leuven University of Antwerp-imec(安特卫普-imec 联合大学KU Leuven);KU Leuven;University of Antwerp-imec(安特卫普-imec 大学)
Adversarial Robustness and Feature Impact Analysis for Driver Drowsiness Detection
Research Group on Intelligent Engineering and Computing for Advanced Innovation and Development (GECAD), School of Engineering, Polytechnic of Porto (ISEP/IPP)(波尔图理工学院工程学院智能工程与先进创新发展计算研究组(GECAD));CardioID Technologies(CardioID科技公司)