SHAPoint: Task-Agnostic, Efficient, and Interpretable Point-Based Risk Scoring via Shapley Values

2025-09-28

AI Summary

Traditional interpretable risk scoring methods rely on manual preprocessing, task-specific modeling, and strong assumptions—limiting flexibility and predictive performance. To address this, we propose SHAPoint: a task-agnostic, efficient point-based risk scoring framework that systematically integrates Shapley values into point-based scoring for the first time. SHAPoint synergizes the high predictive accuracy of gradient-boosted trees with the theoretically grounded, local interpretability of Shapley values. It requires no feature engineering, natively handles missing data and monotonicity constraints, and applies seamlessly to classification, regression, and survival analysis. Experiments demonstrate that SHAPoint matches the predictive accuracy of state-of-the-art models while achieving substantially faster inference. Moreover, it yields concise, compact, and clinically interpretable scoring rules. By unifying rigorous attribution theory with practical risk modeling, SHAPoint establishes a general, robust paradigm for interpretable risk assessment.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Probabilistic Programming
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Interpretable risk scores play a vital role in clinical decision support, yet traditional methods for deriving such scores often rely on manual preprocessing, task-specific modeling, and simplified assumptions that limit their flexibility and predictive power. We present SHAPoint, a novel, task-agnostic framework that integrates the predictive accuracy of gradient boosted trees with the interpretability of point-based risk scores. SHAPoint supports classification, regression, and survival tasks, while also inheriting valuable properties from tree-based models, such as native handling of missing data and support for monotonic constraints. Compared to existing frameworks, SHAPoint offers superior flexibility, reduced reliance on manual preprocessing, and faster runtime performance. Empirical results show that SHAPoint produces compact and interpretable scores with predictive performance comparable to state-of-the-art methods, but at a fraction of the runtime, making it a powerful tool for transparent and scalable risk stratification.

Problem

• Develops task-agnostic interpretable risk scoring via Shapley values
• Integrates gradient boosted trees accuracy with point-based interpretability
• Enables transparent risk stratification with minimal preprocessing and faster runtime

Innovation

• Integrates gradient boosted trees with interpretable risk scores
• Supports classification, regression, and survival analysis tasks
• Provides fast runtime with native missing data handling