BAM: Bivariate Additive Models for Interpretable Tabular Prediction

概要

Generalized Additive Models with pairwise interactions (GA$^2$Ms) are regarded as glass-box models, since each pairwise shape function can be visualized as a contour plot. In practice, the shape functions learned by methods such as EBM or NODE-GAM are so complex (irregular contours and sharp local fluctuations) that a practitioner cannot meaningfully interpret them or check them against domain knowledge. Auditability in this stronger sense requires substantially simpler shape functions. We argue that such simplification can match the accuracy of standard GAMs if the right feature pairs are captured. We propose Bivariate Additive Models (BAMs), a class of GA$^2$Ms in which every pairwise shape function is constrained to a simple parametric form, instantiated as an asymmetric bivariate Gaussian. Each component is visualizable as a unimodal contour map, and its feature pair is selected automatically during training. Experiments on six tabular benchmarks show that BAM attains predictive accuracy comparable to standard GAMs and approaches that of GA$^2$Ms. We also show the trade-off between accuracy and the number of components to inspect.

参考リンク

• Paper ↗

リンク先のライセンス・再現性・データ条件をご確認ください。自動分類や実装の目安は保証ではありません。

実装を始めるプロンプト

参考リンクを確認したうえで、次の内容をコピーして利用できます。

# 論文・記事
BAM: Bivariate Additive Models for Interpretable Tabular Prediction

# 概要
Generalized Additive Models with pairwise interactions (GA$^2$Ms) are regarded as glass-box models, since each pairwise shape function can be visualized as a contour plot. In practice, the shape…

# 原典
https://arxiv.org/abs/2611.05940

上記の手法を実装してください。実行前に依存関係とライセンスを確認してください。