Pathwise Explanation of ReLU Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lim, Seongwoo, Jo, Won, Lee, Joohyung, Choi, Jaesik
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909656450859008
author Lim, Seongwoo
Jo, Won
Lee, Joohyung
Choi, Jaesik
author_facet Lim, Seongwoo
Jo, Won
Lee, Joohyung
Choi, Jaesik
contents Neural networks have demonstrated a wide range of successes, but their ``black box" nature raises concerns about transparency and reliability. Previous research on ReLU networks has sought to unwrap these networks into linear models based on activation states of all hidden units. In this paper, we introduce a novel approach that considers subsets of the hidden units involved in the decision making path. This pathwise explanation provides a clearer and more consistent understanding of the relationship between the input and the decision-making process. Our method also offers flexibility in adjusting the range of explanations within the input, i.e., from an overall attribution input to particular components within the input. Furthermore, it allows for the decomposition of explanations for a given input for more detailed explanations. Experiments demonstrate that our method outperforms others both quantitatively and qualitatively.
format Preprint
id arxiv_https___arxiv_org_abs_2506_18037
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Pathwise Explanation of ReLU Neural Networks
Lim, Seongwoo
Jo, Won
Lee, Joohyung
Choi, Jaesik
Machine Learning
Artificial Intelligence
Neural networks have demonstrated a wide range of successes, but their ``black box" nature raises concerns about transparency and reliability. Previous research on ReLU networks has sought to unwrap these networks into linear models based on activation states of all hidden units. In this paper, we introduce a novel approach that considers subsets of the hidden units involved in the decision making path. This pathwise explanation provides a clearer and more consistent understanding of the relationship between the input and the decision-making process. Our method also offers flexibility in adjusting the range of explanations within the input, i.e., from an overall attribution input to particular components within the input. Furthermore, it allows for the decomposition of explanations for a given input for more detailed explanations. Experiments demonstrate that our method outperforms others both quantitatively and qualitatively.
title Pathwise Explanation of ReLU Neural Networks
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.18037