EviNAM: Intelligibility and Uncertainty via Evidential Neural Additive Models

TL;DR

EviNAM is a novel neural additive model that provides interpretable feature contributions and reliable uncertainty estimates in a single pass, enhancing decision-making in regression tasks.

Contribution

It introduces EviNAM, combining evidential learning with NAMs to deliver both interpretability and uncertainty estimation simultaneously.

Findings

EviNAM achieves state-of-the-art predictive performance.

It accurately estimates both aleatoric and epistemic uncertainty.

The method extends naturally to classification and generalized additive models.

Abstract

Intelligibility and accurate uncertainty estimation are crucial for reliable decision-making. In this paper, we propose EviNAM, an extension of evidential learning that integrates the interpretability of Neural Additive Models (NAMs) with principled uncertainty estimation. Unlike standard Bayesian neural networks and previous evidential methods, EviNAM enables, in a single pass, both the estimation of the aleatoric and epistemic uncertainty as well as explicit feature contributions. Experiments on synthetic and real data demonstrate that EviNAM matches state-of-the-art predictive performance. While we focus on regression, our method extends naturally to classification and generalized additive models, offering a path toward more intelligible and trustworthy predictions.