AI Insight
This study introduces CoGA2M, a multi-resolution spline-based Generalised Additive Model designed for interpretable machine learning on tabular data. The model demonstrates competitive performance with state-of-the-art methods on the TALENT benchmark suite, ranking in the top third among 29-32 baseline models. Critically, when tested under distribution shift conditions where training and test data differ, CoGA2M outperforms LightGBM, revealing that the structured additive approach provides robustness advantages in out-of-distribution scenarios.
Why it matters
This research addresses a significant gap in interpretable machine learning by showing that transparent, structure-based models can compete with black-box methods while offering better robustness when applied to real-world data that differs from training conditions. The findings are particularly relevant for high-stakes applications in healthcare, finance, and policy where both interpretability and reliability under changing conditions are essential.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
Abstract: Generalised Additive Models (GAMs) are a cornerstone of interpretable machine learning, offering transparent predictions through structured additive decompositions. Despite advances from classical GAMs to modern neural and spline-based variants, two notable gaps remain. First, comprehensive benchmarking of spline-based GAMs on large-scale tabular learning tasks, the dominant modality in applied settings, is lacking. Second, their robustness under distribution shift, a setting where structured inductive biases should confer advantages, remains unexplored. We propose a multi-resolution, higher-order Composite Generalised Additive Model (CoGA2M) and evaluate it on the TALENT benchmark suite. CoGA2M ranks in the top third across all baselines (29 for regression, 32 for classification), establishing competitive performance with state-of-the-art tabular methods. We further extend TALENT with out-of-distribution (OOD) splits exhibiting train-test distributional shift and benchmark CoGA2M against LightGBM, a consistently top-performing model on TALENT. Under distribution shift, CoGA2M outperforms LightGBM, demonstrating a clear structural advantage. Our results position spline-based GAMs as strong contenders for tabular prediction and reveal a previously unrecognised benefit of additive structure in the OOD regime.
Source: Rethinking GA2Ms: performance and out-of-distribution generalisation