SAILS: Surrogate-based Analysis of Interactions via Local Effect Smooths
2026-06-08 • Artificial Intelligence
Artificial IntelligenceMachine Learning
AI summaryⓘ
The authors created a method called SAILS to better understand how two features in a machine learning model work together. Unlike other tools that just say if two features interact, their method explains the type of interaction and shows easy-to-understand visualizations. They test their approach on simulated and real data to show it works well for pairs of features but note it struggles with very linked features or interactions involving many features. This helps explain machine learning models in more detail than before.
feature interactionmachine learningexplainable AIgeneralized additive modelssurrogate modelslocal effectsinteraction detectionvisualizationmodel-agnosticsignificance tests
Authors
Timo Heiß, Julia Herbinger, Bernd Bischl, Giuseppe Casalicchio
Abstract
Feature interactions drive much of the predictive power of machine learning models, yet existing explanation methods only detect and quantify interactions without revealing their functional form, or visualize only restricted interaction types. We propose Surrogate-based Analysis of Interactions via Local effect Smooths (SAILS), a model-agnostic framework that analyzes pairwise interactions through interpretable generalized additive model (GAM) surrogates fitted to the local effects of a black-box model. For each interval of a feature of interest, the surrogate smooth terms isolate the interaction components on derivative level, enabling (i) interaction detection through a heuristic derived from significance tests on smooth terms, (ii) interaction form categorization into linear, product-separable, and non-product-separable types, and (iii) tailored, interpretable visualizations for each interaction type. We empirically validate the framework through controlled simulations and a real-world task, demonstrating its effectiveness for pairwise interactions, with limitations under strong feature correlations and higher-order interactions. SAILS fills a notable gap in the XAI toolbox, going beyond detection of interactions alone to characterizing their functional form.