🤖 AI Summary
This work addresses a critical limitation of traditional interpretable models, which often overlook interaction effects that emerge only through specific joint configurations of variables, leading to significant information loss. To overcome this, the authors propose the IAIML framework—the first approach to explicitly model pairwise interactions within compact interpretable models. IAIML identifies key interactions via adaptive discretization, finite-grid interaction scoring, and an interaction-aware feature admission mechanism, while controlling model complexity through an interpretability budget. Evaluated across 40 datasets, IAIML achieves an average AUC only 1.4% lower than gradient-boosted trees, with interpretability components reduced by 14–28 times. Notably, it substantially outperforms existing baselines in scenarios dominated by strong interactions, effectively balancing predictive accuracy and model interpretability.
📝 Abstract
Inherently interpretable classifiers for tabular data typically rely on sparse features, rules, or patterns that users can inspect directly. The marginal feature-screening step common to these methods can discard variables whose predictive value emerges only through joint configurations with other variables. We present Interaction Aware Interpretable Machine Learning (IAIML), a framework that addresses this limitation through three coordinated mechanisms: adaptive per-feature discretization, finite-grid pairwise interaction scoring, and a partitioned explanation budget. Detected interactions are routed through one of two strategies: relaxing the screening filter so that interaction-supported variables enter the pattern search, or constructing explicit pair terms for a sparse downstream classifier. On a 40-dataset panel comprising 24 real-world tabular benchmarks and 16 synthetic interaction stress tests, evaluated under nested cross-validation, IAIML achieves mean AUC within 1.4 points of tuned gradient-boosted ensembles while requiring roughly 14--28 times fewer fitted explanation components. On datasets with strong pairwise interaction structure and low marginal signal, IAIML outperforms all baselines. Among compact interpretable methods, IAIML is comparable to RuleFit in AUC and component count and is less expensive to tune. EBM obtains a small but significant AUC advantage across the full panel, with a substantially larger lookup-table footprint. Performance degrades on datasets requiring higher-order interactions beyond the pairwise scope. Component-isolated ablations confirm that adaptive discretization and interaction-aware admission each contribute incrementally. These results support IAIML as a compact, interaction-aware framework appropriate for settings where bounded explanation size and controlled treatment of feature interactions are design requirements.