A novel post-hoc explanation comparison metric and applications

Shreyan Mitra, Leilani Gilpin
11/17/2023
cs.LGcs.AI

Official code: 11301858/XAISuite (7 stars, last commit 2024-04-21).

Abstract

Explanatory systems make the behavior of machine learning models more transparent, but are often inconsistent. To quantify the differences between explanatory systems, this paper presents the Shreyan Distance, a novel metric based on the weighted difference between ranked feature importance lists produced by such systems. This paper uses the Shreyan Distance to compare two explanatory systems, SHAP and LIME, for both regression and classification learning tasks. Because we find that the average Shreyan Distance varies significantly between these two tasks, we conclude that consistency between explainers not only depends on inherent properties of the explainers themselves, but also the type of learning task. This paper further contributes the XAISuite library, which integrates the Shreyan distance algorithm into machine learning pipelines.

View on arXivView PDF

Code Implementations(1)

11301858/XAISuiteOfficial

XAISuite: Train machine learning models, generate explanations, and compare different explanation systems with just a simple line of code.

71MATLAB, PythonOct 23, 20222 years agoNOASSERTION
comparison-toolexplainable-aiexplainable-machine-learninginterpretable-machine-learningmachine-learning+6 more

Similar Papers

The XAISuite framework and the implications of explanatory system dissonance

87% match1 repo72023

An Empirical Evaluation of the Rashomon Effect in Explainable Machine Learning

83% match2023

Unified Explanations in Machine Learning Models: A Perturbation Approach

83% match1 repo2024

Fast and Accurate Explanations of Distance-Based Classifiers by Uncovering Latent Explanatory Structures

83% match2025

The Disagreement Problem in Faithfulness Metrics

82% match2023

A Perspective on Explainable Artificial Intelligence Methods: SHAP and LIME

81% match2023

Cite this paper

@article{mitra2023novel,
  title  = {A novel post-hoc explanation comparison metric and applications},
  author = {Shreyan Mitra and Leilani Gilpin},
  year   = {2023},
  eprint = {2311.10811},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2311.10811v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2311.10811.svg)](https://codeofpaper.com/paper/2311.10811)

Discussion