Guardado en:
Detalles Bibliográficos
Autores principales: Buchnajzer, Zuzanna, Dobek, Kacper, Hapke, Stanisław, Jankowski, Daniel, Krawiec, Krzysztof
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2411.12070
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917323122671616
author Buchnajzer, Zuzanna
Dobek, Kacper
Hapke, Stanisław
Jankowski, Daniel
Krawiec, Krzysztof
author_facet Buchnajzer, Zuzanna
Dobek, Kacper
Hapke, Stanisław
Jankowski, Daniel
Krawiec, Krzysztof
contents Deep learning architectures based on convolutional neural networks tend to rely on continuous, smooth features. While this characteristics provides significant robustness and proves useful in many real-world tasks, it is strikingly incompatible with the physical characteristic of the world, which, at the scale in which humans operate, comprises crisp objects, typically representing well-defined categories. This study proposes a class of neurosymbolic systems that learn by reconstructing images in terms of visual primitives and are thus forced to form high-level, structural explanations of them. When applied to the task of diagnosing abnormalities in histological imaging, the method proved superior to a conventional deep learning architecture in terms of classification accuracy, while being more transparent.
format Preprint
id arxiv_https___arxiv_org_abs_2411_12070
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Autoassociative Learning of Structural Representations for Modeling and Classification in Medical Imaging
Buchnajzer, Zuzanna
Dobek, Kacper
Hapke, Stanisław
Jankowski, Daniel
Krawiec, Krzysztof
Computer Vision and Pattern Recognition
Machine Learning
68T05
I.2; I.2.6; I.2.10
Deep learning architectures based on convolutional neural networks tend to rely on continuous, smooth features. While this characteristics provides significant robustness and proves useful in many real-world tasks, it is strikingly incompatible with the physical characteristic of the world, which, at the scale in which humans operate, comprises crisp objects, typically representing well-defined categories. This study proposes a class of neurosymbolic systems that learn by reconstructing images in terms of visual primitives and are thus forced to form high-level, structural explanations of them. When applied to the task of diagnosing abnormalities in histological imaging, the method proved superior to a conventional deep learning architecture in terms of classification accuracy, while being more transparent.
title Autoassociative Learning of Structural Representations for Modeling and Classification in Medical Imaging
topic Computer Vision and Pattern Recognition
Machine Learning
68T05
I.2; I.2.6; I.2.10
url https://arxiv.org/abs/2411.12070