LocalGLMnet: interpretable deep learning for tabular data

TL;DR

LocalGLMnet introduces an interpretable deep learning architecture for tabular data that combines the transparency of generalized linear models with the predictive power of deep learning, enabling variable selection and interpretability.

Contribution

It proposes a novel neural network architecture inspired by generalized linear models that offers improved interpretability and variable selection for tabular data.

Findings

Provides an additive decomposition similar to Shapley values and integrated gradients.

Achieves superior predictive performance compared to traditional models.

Enables variable selection within a deep learning framework.

Abstract

Deep learning models have gained great popularity in statistical modeling because they lead to very competitive regression models, often outperforming classical statistical models such as generalized linear models. The disadvantage of deep learning models is that their solutions are difficult to interpret and explain, and variable selection is not easily possible because deep learning models solve feature engineering and variable selection internally in a nontransparent way. Inspired by the appealing structure of generalized linear models, we propose a new network architecture that shares similar features as generalized linear models, but provides superior predictive power benefiting from the art of representation learning. This new architecture allows for variable selection of tabular data and for interpretation of the calibrated deep learning model, in fact, our approach provides an additive decomposition in the spirit of Shapley values and integrated gradients.