SHAPNN: Shapley Value Regularized Tabular Neural Network

TL;DR

SHAPNN is a deep neural network architecture for tabular data that incorporates Shapley value regularization to improve performance, explainability, and continual learning without additional computational costs.

Contribution

It introduces a novel regularization technique using real-time Shapley value estimates in deep neural networks for tabular data, enhancing accuracy and interpretability.

Findings

SHAPNN outperforms state-of-the-art models in AUROC.

It provides valid explanations with no extra computational overhead.

The method improves robustness to streaming data.

Abstract

We present SHAPNN, a novel deep tabular data modeling architecture designed for supervised learning. Our approach leverages Shapley values, a well-established technique for explaining black-box models. Our neural network is trained using standard backward propagation optimization methods, and is regularized with realtime estimated Shapley values. Our method offers several advantages, including the ability to provide valid explanations with no computational overhead for data instances and datasets. Additionally, prediction with explanation serves as a regularizer, which improves the model's performance. Moreover, the regularized prediction enhances the model's capability for continual learning. We evaluate our method on various publicly available datasets and compare it with state-of-the-art deep neural network models, demonstrating the superior performance of SHAPNN in terms of AUROC, transparency, as well as robustness to streaming data.