AI RESEARCH
Domain-informed explainable boosting machines for trustworthy lateral spread predictions
arXiv CS.LG
•
ArXi:2603.17175v1 Announce Type: new Explainable Boosting Machines (EBMs) provide transparent predictions through additive shape functions, enabling direct inspection of feature contributions. However, EBMs can learn non-physical relationships that reduce their reliability in natural hazard applications. This study presents a domain-informed framework to improve the physical consistency of EBMs for lateral spreading prediction. Our approach modifies learned shape functions based on domain knowledge. These modifications correct non-physical behavior while maintaining data-driven patterns.