neuralGAM: An R Package for Fitting Generalized Additive Neural Networks

TL;DR

neuralGAM is an R package that combines neural networks with generalized additive models to create interpretable and accurate deep learning models, addressing the black-box problem.

Contribution

This paper introduces neuralGAM, an R package that implements a flexible framework for fitting interpretable generalized additive neural networks.

Findings

Provides a highly interpretable neural network model.

Demonstrates effectiveness on synthetic and real data.

No restrictions on neural network architecture.

Abstract

Nowadays, Neural Networks are considered one of the most effective methods for various tasks such as anomaly detection, computer-aided disease detection, or natural language processing. However, these networks suffer from the ``black-box'' problem which makes it difficult to understand how they make decisions. In order to solve this issue, an R package called neuralGAM is introduced. This package implements a Neural Network topology based on Generalized Additive Models, allowing to fit an independent Neural Network to estimate the contribution of each feature to the output variable, yielding a highly accurate and interpretable Deep Learning model. The neuralGAM package provides a flexible framework for training Generalized Additive Neural Networks, which does not impose any restrictions on the Neural Network architecture. We illustrate the use of the neuralGAM package in both synthetic and real data examples.