Bound Propagation meets Constraint Simplification: Improving Logic-based XAI for Neural Networks

TL;DR

This paper enhances logic-based neural network explanations by combining bound propagation with constraint simplification, significantly reducing computation time while maintaining formal guarantees.

Contribution

It introduces a novel method that integrates bound propagation with constraint simplification to improve explanation efficiency for neural networks.

Findings

Explanation time reduced by up to 89.26%

Method scales better for larger networks

Maintains formal correctness guarantees

Abstract

Logic-based methods for explaining neural network decisions offer formal guarantees of correctness and non-redundancy, but they often suffer from high computational costs, especially for large networks. In this work, we improve the efficiency of such methods by combining bound propagation with constraint simplification. These simplifications, derived from the propagation, tighten neuron bounds and eliminate unnecessary binary variables, making the explanation process more efficient. Our experiments suggest that combining these techniques reduces explanation time by up to 89.26\%, particularly for larger neural networks.