卷积神经网络的可解释性和细粒度视觉解释

Jörg Wagner, Jan M. Köhler, Tobias Gindele, Leon Hetzel, Jakob Thaddäus Wiedemer, Sven Behnke
{"title":"卷积神经网络的可解释性和细粒度视觉解释","authors":"Jörg Wagner, Jan M. Köhler, Tobias Gindele, Leon Hetzel, Jakob Thaddäus Wiedemer, Sven Behnke","doi":"10.1109/CVPR.2019.00931","DOIUrl":null,"url":null,"abstract":"To verify and validate networks, it is essential to gain insight into their decisions, limitations as well as possible shortcomings of training data. In this work, we propose a post-hoc, optimization based visual explanation method, which highlights the evidence in the input image for a specific prediction. Our approach is based on a novel technique to defend against adversarial evidence (i.e. faulty evidence due to artefacts) by filtering gradients during optimization. The defense does not depend on human-tuned parameters. It enables explanations which are both fine-grained and preserve the characteristics of images, such as edges and colors. The explanations are interpretable, suited for visualizing detailed evidence and can be tested as they are valid model inputs. We qualitatively and quantitatively evaluate our approach on a multitude of models and datasets.","PeriodicalId":6711,"journal":{"name":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"33 1","pages":"9089-9099"},"PeriodicalIF":0.0000,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"99","resultStr":"{\"title\":\"Interpretable and Fine-Grained Visual Explanations for Convolutional Neural Networks\",\"authors\":\"Jörg Wagner, Jan M. Köhler, Tobias Gindele, Leon Hetzel, Jakob Thaddäus Wiedemer, Sven Behnke\",\"doi\":\"10.1109/CVPR.2019.00931\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"To verify and validate networks, it is essential to gain insight into their decisions, limitations as well as possible shortcomings of training data. In this work, we propose a post-hoc, optimization based visual explanation method, which highlights the evidence in the input image for a specific prediction. Our approach is based on a novel technique to defend against adversarial evidence (i.e. faulty evidence due to artefacts) by filtering gradients during optimization. The defense does not depend on human-tuned parameters. It enables explanations which are both fine-grained and preserve the characteristics of images, such as edges and colors. The explanations are interpretable, suited for visualizing detailed evidence and can be tested as they are valid model inputs. We qualitatively and quantitatively evaluate our approach on a multitude of models and datasets.\",\"PeriodicalId\":6711,\"journal\":{\"name\":\"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)\",\"volume\":\"33 1\",\"pages\":\"9089-9099\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2019-06-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"99\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/CVPR.2019.00931\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CVPR.2019.00931","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 99

摘要

为了验证和验证网络,必须深入了解它们的决策、限制以及训练数据可能存在的缺点。在这项工作中,我们提出了一种基于事后优化的视觉解释方法,该方法突出显示输入图像中的证据以进行特定的预测。我们的方法是基于一种新的技术,通过在优化过程中过滤梯度来防御对抗性证据(即由于伪影导致的错误证据)。防御不依赖于人为调整的参数。它使解释既精细又保留图像的特征,如边缘和颜色。这些解释是可解释的,适合于可视化详细的证据,并且可以作为有效的模型输入进行测试。我们在大量模型和数据集上定性和定量地评估我们的方法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Interpretable and Fine-Grained Visual Explanations for Convolutional Neural Networks
To verify and validate networks, it is essential to gain insight into their decisions, limitations as well as possible shortcomings of training data. In this work, we propose a post-hoc, optimization based visual explanation method, which highlights the evidence in the input image for a specific prediction. Our approach is based on a novel technique to defend against adversarial evidence (i.e. faulty evidence due to artefacts) by filtering gradients during optimization. The defense does not depend on human-tuned parameters. It enables explanations which are both fine-grained and preserve the characteristics of images, such as edges and colors. The explanations are interpretable, suited for visualizing detailed evidence and can be tested as they are valid model inputs. We qualitatively and quantitatively evaluate our approach on a multitude of models and datasets.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信