Salvato in:
Dettagli Bibliografici
Autori principali: Xuan, Xiwei, Deng, Ziquan, Lin, Hsuan-Tien, Kong, Zhaodan, Ma, Kwan-Liu
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:https://arxiv.org/abs/2303.00244
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910459323482112
author Xuan, Xiwei
Deng, Ziquan
Lin, Hsuan-Tien
Kong, Zhaodan
Ma, Kwan-Liu
author_facet Xuan, Xiwei
Deng, Ziquan
Lin, Hsuan-Tien
Kong, Zhaodan
Ma, Kwan-Liu
contents Researchers have proposed various methods for visually interpreting the Convolutional Neural Network (CNN) via saliency maps, which include Class-Activation-Map (CAM) based approaches as a leading family. However, in terms of the internal design logic, existing CAM-based approaches often overlook the causal perspective that answers the core "why" question to help humans understand the explanation. Additionally, current CNN explanations lack the consideration of both necessity and sufficiency, two complementary sides of a desirable explanation. This paper presents a causality-driven framework, SUNY, designed to rationalize the explanations toward better human understanding. Using the CNN model's input features or internal filters as hypothetical causes, SUNY generates explanations by bi-directional quantifications on both the necessary and sufficient perspectives. Extensive evaluations justify that SUNY not only produces more informative and convincing explanations from the angles of necessity and sufficiency, but also achieves performances competitive to other approaches across different CNN architectures over large-scale datasets, including ILSVRC2012 and CUB-200-2011.
format Preprint
id arxiv_https___arxiv_org_abs_2303_00244
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle SUNY: A Visual Interpretation Framework for Convolutional Neural Networks from a Necessary and Sufficient Perspective
Xuan, Xiwei
Deng, Ziquan
Lin, Hsuan-Tien
Kong, Zhaodan
Ma, Kwan-Liu
Computer Vision and Pattern Recognition
Artificial Intelligence
Researchers have proposed various methods for visually interpreting the Convolutional Neural Network (CNN) via saliency maps, which include Class-Activation-Map (CAM) based approaches as a leading family. However, in terms of the internal design logic, existing CAM-based approaches often overlook the causal perspective that answers the core "why" question to help humans understand the explanation. Additionally, current CNN explanations lack the consideration of both necessity and sufficiency, two complementary sides of a desirable explanation. This paper presents a causality-driven framework, SUNY, designed to rationalize the explanations toward better human understanding. Using the CNN model's input features or internal filters as hypothetical causes, SUNY generates explanations by bi-directional quantifications on both the necessary and sufficient perspectives. Extensive evaluations justify that SUNY not only produces more informative and convincing explanations from the angles of necessity and sufficiency, but also achieves performances competitive to other approaches across different CNN architectures over large-scale datasets, including ILSVRC2012 and CUB-200-2011.
title SUNY: A Visual Interpretation Framework for Convolutional Neural Networks from a Necessary and Sufficient Perspective
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2303.00244