Unbiased Attribution with Intrinsic Information

Zhiyu Zhu, Zhibo Jin, Jiayu Zhang, Jianlong Zhou, Fang Chen

Keywords: Interpretability, Attribution

Abstract: The importance of attribution algorithms in the AI field lies in enhancing model transparency, diagnosing and improving models, ensuring fairness, and increasing user understanding. Gradient-based attribution methods have become the most critical because of their high computational efficiency, continuity, wide applicability, and flexibility. However, current gradient-based attribution algorithms require the introduction of additional class information to interpret model decisions, which can lead to issues of information ignorance and extra information. Information ignorance can obscure important features relevant to the current model decision, while extra information introduces irrelevant data that can cause feature leakage in the attribution process. To address these issues, we propose the Attribution with Intrinsic Information (AII) algorithm, which analyzes model decisions without the need for specified class information. Additionally, to better evaluate the potential of current attribution algorithms, we introduce the metrics of insertion confusion and deletion confusion alongside existing mainstream metrics. To continuously advance research in the field of explainable AI (XAI), our algorithm is open-sourced at https://anonymous.4open.science/r/AII-787D/.

Primary Area: interpretability and explainable AI

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 9383