TY - GEN
T1 - Towards Better Visualizing the Decision Basis of Networks via Unfold and Conquer Attribution Guidance
AU - Hong, Jung Ho
AU - Nam, Woo Jeoung
AU - Jeon, Kyu Sung
AU - Lee, Seong Whan
N1 - Publisher Copyright:
Copyright © 2023, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
PY - 2023/6/27
Y1 - 2023/6/27
N2 - Revealing the transparency of Deep Neural Networks (DNNs) has been widely studied to describe the decision mechanisms of network inner structures. In this paper, we propose a novel post-hoc framework, Unfold and Conquer Attribution Guidance (UCAG), which enhances the explainability of the network decision by spatially scrutinizing the input features with respect to the model confidence. Addressing the phenomenon of missing detailed descriptions, UCAG sequentially complies with the confidence of slices of the image, leading to providing an abundant and clear interpretation. Therefore, it is possible to enhance the representation ability of explanation by preserving the detailed descriptions of assistant input features, which are commonly overwhelmed by the main meaningful regions. We conduct numerous evaluations to validate the performance in several metrics: i) deletion and insertion, ii) (energy-based) pointing games, and iii) positive and negative density maps. Experimental results, including qualitative comparisons, demonstrate that our method outperforms the existing methods with the nature of clear and detailed explanations and applicability.
AB - Revealing the transparency of Deep Neural Networks (DNNs) has been widely studied to describe the decision mechanisms of network inner structures. In this paper, we propose a novel post-hoc framework, Unfold and Conquer Attribution Guidance (UCAG), which enhances the explainability of the network decision by spatially scrutinizing the input features with respect to the model confidence. Addressing the phenomenon of missing detailed descriptions, UCAG sequentially complies with the confidence of slices of the image, leading to providing an abundant and clear interpretation. Therefore, it is possible to enhance the representation ability of explanation by preserving the detailed descriptions of assistant input features, which are commonly overwhelmed by the main meaningful regions. We conduct numerous evaluations to validate the performance in several metrics: i) deletion and insertion, ii) (energy-based) pointing games, and iii) positive and negative density maps. Experimental results, including qualitative comparisons, demonstrate that our method outperforms the existing methods with the nature of clear and detailed explanations and applicability.
UR - http://www.scopus.com/inward/record.url?scp=85168239578&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85168239578
T3 - Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023
SP - 7884
EP - 7892
BT - AAAI-23 Technical Tracks 7
A2 - Williams, Brian
A2 - Chen, Yiling
A2 - Neville, Jennifer
PB - AAAI press
T2 - 37th AAAI Conference on Artificial Intelligence, AAAI 2023
Y2 - 7 February 2023 through 14 February 2023
ER -