Four Axiomatic Characterizations of the Integrated Gradients Attribution Method

Authors: Daniel Lundstrom, Meisam Razaviyayn

JMLR 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We present four axiomatic characterizations of IG, establishing IG as the unique method satisfying four different sets of axioms. Keywords: Machine Learning Explainability, Attribution Methods, Axiomatic Approach, Integrated Gradients, Path Methods
Researcher Affiliation Academia Daniel Lundstrom EMAIL Department of Mathematics University of Southern California Los Angeles, CA 90007, USA Meisam Razaviyayn EMAIL Departments of Industrial and Systems Eng., Electrical Eng., and Computer Science University of Southern California Los Angeles, CA 90007, USA
Pseudocode No The paper does not contain any sections or figures explicitly labeled as 'Pseudocode' or 'Algorithm', nor does it present structured code-like blocks for procedures.
Open Source Code No The paper includes a license for the document itself ('License: CC-BY 4.0, see https://creativecommons.org/licenses/by/4.0/. Attribution requirements are provided at http://jmlr.org/papers/v26/23-0671.html.'), but it does not contain any statements about making the source code for the described methodology publicly available, nor does it provide links to any code repositories or mention code in supplementary materials.
Open Datasets No This paper is theoretical, focusing on axiomatic characterizations. It discusses machine learning models and attribution methods in general terms, referencing fields like computer vision and natural language processing, but does not conduct experiments using specific datasets or provide access information for any open datasets.
Dataset Splits No This paper is theoretical and does not involve experiments on datasets, therefore, it does not provide any information regarding training/test/validation dataset splits.
Hardware Specification No This paper is theoretical and presents axiomatic characterizations and proofs. It does not describe any experimental setup or the hardware used to run experiments.
Software Dependencies No This paper is theoretical and focuses on mathematical characterizations. It does not describe any experimental implementation or list specific software dependencies with version numbers.
Experiment Setup No This paper is theoretical and presents axiomatic characterizations of a method. It does not include any experimental setup details, hyperparameters, or training configurations.