Statistical Inference for Explainable Boosting Machines
This work addresses the need for reliable uncertainty estimates in explainable AI models, particularly for practitioners using EBMs, though it is incremental as it builds on existing gradient boosting and regularization techniques.
The paper tackled the problem of uncertainty quantification in Explainable Boosting Machines (EBMs), which previously required computationally intensive bootstrapping, by deriving statistical inference methods with theoretical guarantees, achieving minimax-optimal mean squared error rates of O(pn^{-2/3}) for Lipschitz GAMs and enabling efficient confidence intervals.
Explainable boosting machines (EBMs) are popular "glass-box" models that learn a set of univariate functions using boosting trees. These achieve explainability through visualizations of each feature's effect. However, unlike linear model coefficients, uncertainty quantification for the learned univariate functions requires computationally intensive bootstrapping, making it hard to know which features truly matter. We provide an alternative using recent advances in statistical inference for gradient boosting, deriving methods for statistical inference as well as end-to-end theoretical guarantees. Using a moving average instead of a sum of trees (Boulevard regularization) allows the boosting process to converge to a feature-wise kernel ridge regression. This produces asymptotically normal predictions that achieve the minimax-optimal mean squared error for fitting Lipschitz GAMs with $p$ features at rate $O(pn^{-2/3})$, successfully avoiding the curse of dimensionality. We then construct prediction intervals for the response and confidence intervals for each learned univariate function with a runtime independent of the number of datapoints, enabling further explainability within EBMs.