Generating attribution maps with disentangled masked backpropagation
10.1109/ICCV48922.2021.00094
Inclou dades d'ús des de 2022
Cita com:
hdl:2117/385247
Tipus de documentText en actes de congrés
Data publicació2021
EditorInstitute of Electrical and Electronics Engineers (IEEE)
Condicions d'accésAccés obert
Tots els drets reservats. Aquesta obra està protegida pels drets de propietat intel·lectual i
industrial corresponents. Sense perjudici de les exempcions legals existents, queda prohibida la seva
reproducció, distribució, comunicació pública o transformació sense l'autorització del titular dels drets
Abstract
Attribution map visualization has arisen as one of the most effective techniques to understand the underlying inference process of Convolutional Neural Networks. In this task, the goal is to compute an score for each image pixel related to its contribution to the network output. In this paper, we introduce Disentangled Masked Backpropagation (DMBP), a novel gradient-based method that leverages on the piecewise linear nature of ReLU networks to decompose the model function into different linear mappings. This decomposition aims to disentangle the attribution maps into positive, negative and nuisance factors by learning a set of variables masking the contribution of each filter during back-propagation. A thorough evaluation over standard architectures (ResNet50 and VGG16) and benchmark datasets (PASCAL VOC and ImageNet) demonstrates that DMBP generates more visually interpretable attribution maps than previous approaches. Additionally, we quantitatively show that the maps produced by our method are more consistent with the true contribution of each pixel to the final network output.
Descripció
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
CitacióRuiz, A.; Agudo, A.; Moreno-Noguer, F. Generating attribution maps with disentangled masked backpropagation. A: IEEE International Conference on Computer Vision. "2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition: Virtual, 19-25 June 2021: proceedings". Institute of Electrical and Electronics Engineers (IEEE), 2021, p. 885-894. ISBN 978-1-6654-4509-2. DOI 10.1109/ICCV48922.2021.00094.
ISBN978-1-6654-4509-2
Versió de l'editorhttps://ieeexplore.ieee.org/document/9710348
Fitxers | Descripció | Mida | Format | Visualitza |
---|---|---|---|---|
2472-Generating ... Masked-Backpropagation.pdf | 7,232Mb | Visualitza/Obre |