Rethinking GA2Ms: performance and out-of-distribution generalisation
Organizations: Computer Science and Electronic Engineering Department of the University of Essex · School of Engineering at the University of Edinburgh · Causality in Healthcare AI Hub (CHAI) · Institute for Social and Economic Research and the ESRC Research Centre on Micro-Social Change (MiSoC), Univesrity of Essex
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.
Figures & tables
| Models’ differences and number of OOD cuts | ||||
|---|---|---|---|---|
| Dataset | OOD - ID | LGBM OOD - ID | N | |
| 3D-Estimation-using-RSSI-of-WLAN-dataset-complete-1-target | -0.164 | -0.075 | -0.089 | 13 |
| Another-Dataset-on-used-Fiat-500-(1538-rows) | -0.137 | -0.017 | -0.120 | 12 |
| Bias-correction-r | -0.872 | -0.198 | -0.674 | 34 |
| Brazilian-houses-reproduced | -0.046 | -0.164 | 0.118 | 11 |
| IEEE80211aa-GATS | -0.397 | -0.470 | 0.073 | 49 |
Appendix figures & tables8 assets
Supplementary material from the paper’s appendix.
Appendix
| Models’ for ID and OOD datasets | |||||
|---|---|---|---|---|---|
| Dataset | LGBM | LGBM | OOD generator | ||
| 3D-Estimation-using-RSSI-of-WLAN-dataset-complete-1-target | 0.995 | 0.999 | 0.831 | 0.925 | kNNR |
| Another-Dataset-on-used-Fiat-500-(1538-rows) | 0.853 | 0.828 | 0.730 | 0.813 | kNNR |
| Bias-correction-r | 0.924 | 0.933 | 0.111 | 0.748 | SVR |
| Brazilian-houses-reproduced | 0.988 | 0.989 | 0.944 | 0.831 | kNNR |
| IEEE80211aa-GATS | 0.994 | 0.979 | 0.635 | 0.554 | kNNR |
Tabular ML
Tabular machine learning develops models and tools for learning from structured data arranged in rows and columns.
See all 125 papers