A Robust Unsupervised Ensemble of Feature-Based Explanations using Restricted Boltzmann Machines

TL;DR

This paper introduces a robust ensemble method using Restricted Boltzmann Machines to aggregate feature explanations from multiple algorithms, improving interpretability of deep neural networks.

Contribution

It presents a novel RBM-based ensemble approach that enhances the reliability of feature attribution explanations in deep learning models.

Findings

RBM ensemble outperforms individual attribution methods

The approach yields more consistent explanations across hyperparameters

Experiments on real-world datasets validate the method's effectiveness

Abstract

Understanding the results of deep neural networks is an essential step towards wider acceptance of deep learning algorithms. Many approaches address the issue of interpreting artificial neural networks, but often provide divergent explanations. Moreover, different hyperparameters of an explanatory method can lead to conflicting interpretations. In this paper, we propose a technique for aggregating the feature attributions of different explanatory algorithms using Restricted Boltzmann Machines (RBMs) to achieve a more reliable and robust interpretation of deep neural networks. Several challenging experiments on real-world datasets show that the proposed RBM method outperforms popular feature attribution methods and basic ensemble techniques.