Diffusion Models for Attribution
Files
(Published version)
Date
2025
Authors
Chen, X.
Li, J.
Liu, J.
Liu, L.
Peters, S.
Le, T.D.
Gao, W.
Du, X.
Walsh, A.
Editors
Walsh, T.
Shah, J.
Kolter, Z.
Shah, J.
Kolter, Z.
Advisors
Journal Title
Journal ISSN
Volume Title
Type:
Conference paper
Citation
Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence, 2025 / Walsh, T., Shah, J., Kolter, Z. (ed./s), vol.39, iss.2, pp.2266-2274
Statement of Responsibility
Conference Name
39th AAAI Conference on Artificial Intelligence (25 Feb 2025 - 4 Mar 2025 : PA, Philadelphia)
Abstract
<jats:p>In high-stakes domains such as healthcare, finance, and law, the need for explainable AI is critical. Traditional methods for generating attribution maps, including white-box approaches relying on gradients and black-box techniques that perturb inputs, face challenges like gradient vanishing, blurred attributions, and computational inefficiencies. To overcome these limitations, we introduce a novel approach that leverages diffusion models within the framework of Information Bottleneck (IB) theory. By utilizing the Gaussian noise from diffusion models, we connect the information bottleneck with the Minimum Mean Squared Error (MMSE) from classical information theory, enabling precise calculation of mutual information. This connection leads to a new loss function that minimizes the Signal-to-Noise Ratio (SNR), facilitating efficient optimization and producing high-resolution, pixel-level attribution maps. Our method achieves greater clarity and accuracy in attributions than existing techniques, requiring significantly fewer pixel values to reach the necessary predictive confidence. This work demonstrates the power of diffusion models in advancing explainable AI, particularly in identifying critical input features with high precision.</jats:p>
School/Discipline
Dissertation Note
Provenance
Description
Access Status
Rights
Copyright 2025 Association for the Advancement of Artificial Intelligence