Tag
An arXiv paper formulates payroll fraud detection as inductive link prediction on synthetic HR/ERP graphs, comparing analyst rules, path features, and GNNs. A gradient-boosted model on chained-join features outperformed a GraIL-style GNN on link AP, while the relational entity GNN led under failed identifier joins, showing multi-hop gains are not GNN-specific.
一篇提出 AI 驱动、时序感知的芯片后布线功耗优化框架的论文，使用梯度提升模型集成对功耗修复编辑和修复动作进行排序，同时由确定性 EDA 分析把关约束。论文给出契约与预注册评估协议，但尚未报告训练模型的结果。
This arXiv paper fuses SEC financial data with EPA greenhouse gas registries, using gradient boosting and Mondrian Conformal Prediction to build an algorithmic baseline for corporate emissions and quantify greenwashing via a Conformal-Weighted Continuous Divergence (CWCD) metric, finding that emissions divergence negatively predicts market valuation and profitability.
Microsoft Research researchers developed an end-to-end machine learning pipeline that forecasts geomagnetically induced current risk for 66,935 U.S. substations 30–60 minutes before space-weather events, combining solar-wind forecasts (AE/Dst indices), physics-informed constraints, and local geological conductivity. The system detected nearly 80% of major events during evaluation.
Researchers from Allora Foundation and Liverpool John Moores University introduce the CZAR (Composite Zero-Agnostic Return) loss, a convex, asymmetric, piecewise-quadratic objective tailored for financial log-return prediction that prevents models from collapsing to zero-forecast behavior, and show improved directional accuracy on BTC intraday returns with LightGBM.
CRISP is a scalable coreset method for imbalanced tabular learning that efficiently reduces dataset size while maintaining high accuracy in tasks like fraud detection.
This paper introduces a machine learning framework for estimating pedestrian volumes using GIS-derived built-environment features, achieving better accuracy than traditional models like Negative Binomial GLM through gradient boosting and feature selection.
The paper introduces a framework for integrating explainable AI into CRM systems for customer churn prediction in telecommunications, benchmarking classifiers and using SHAP and LIME for interpretable predictions to enhance retention strategies.
The paper examines XGBoost's sensitivity to outliers in regression tasks and introduces robust loss functions based on M-, S-, and τ-estimators, proposing MM-XGBoost for a better trade-off between robustness and predictive accuracy.
The paper investigates whether guideline-based categorical encodings of continuous predictors can replace continuous inputs in stroke outcome prediction models without sacrificing accuracy, finding comparable performance in most treatment cohorts.
path_boost is a Python package implementing PathBoost, a gradient boosting algorithm for interpretable graph-level prediction. It automatically discovers predictive labeled paths in graphs, supports regression and binary classification, and is compatible with scikit-learn.
A Kaggle Grandmaster is offering a free course covering the entire machine learning process, including EDA, decision trees, gradient boosting, feature engineering, time series, and competition strategies.
This paper investigates when activation steering succeeds or fails for LLMs by analyzing early decoding dynamics. The authors introduce ASTEER, a large testbed of steered generations, and train a GBDT classifier to predict steering outcomes from early hidden states, enabling efficient steering strength search.
This paper presents LiverRisk, a machine learning framework for NAFLD risk prediction that combines gradient-boosted decision trees with conformal prediction to provide calibrated, distribution-free coverage guarantees on individual risk estimates, achieving high AUROC on internal and external cohorts.
This paper introduces a method to predict activation steering effectiveness in language models from early decoding states using a Gradient Boosting Decision Trees (GBDT) classifier, enabling efficient steering strength optimization without full rollouts.
This paper presents a hybrid architecture combining FT-Transformer with gradient-boosted trees via calibration-aware stacking for customer churn prediction on structured tabular data, achieving improved F1 and AUC-ROC on a public bank churn dataset.
A blog post from Flyback demonstrates how a LightGBM feature that ranked #1 in importance actually worsened predictions due to target encoding leakage, highlighting the danger of relying solely on feature importance metrics.
This paper introduces PathBoost, a gradient tree boosting method for graph-level prediction that uses path-based features to compete with graph neural networks while offering better interpretability.