Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Tawalbeh, Saja, Oramas, José
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2305.05349
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909425897308160
author Tawalbeh, Saja
Oramas, José
author_facet Tawalbeh, Saja
Oramas, José
contents Capsule Networks (CapsNets) have been re-introduced as a more compact and interpretable alternative to standard deep neural networks. While recent efforts have proved their compression capabilities, to date, their interpretability properties have not been fully assessed. Here, we conduct a systematic and principled study towards assessing the interpretability of these types of networks. Moreover, we pay special attention towards analyzing the level to which part-whole relationships are indeed encoded within the learned representation. Our analysis in the MNIST, SVHN, PASCAL-part and CelebA datasets suggest that the representations encoded in CapsNets might not be as disentangled nor strictly related to parts-whole relationships as is commonly stated in the literature.
format Preprint
id arxiv_https___arxiv_org_abs_2305_05349
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Towards the Characterization of Representations Learned via Capsule-based Network Architectures
Tawalbeh, Saja
Oramas, José
Machine Learning
Computer Vision and Pattern Recognition
Capsule Networks (CapsNets) have been re-introduced as a more compact and interpretable alternative to standard deep neural networks. While recent efforts have proved their compression capabilities, to date, their interpretability properties have not been fully assessed. Here, we conduct a systematic and principled study towards assessing the interpretability of these types of networks. Moreover, we pay special attention towards analyzing the level to which part-whole relationships are indeed encoded within the learned representation. Our analysis in the MNIST, SVHN, PASCAL-part and CelebA datasets suggest that the representations encoded in CapsNets might not be as disentangled nor strictly related to parts-whole relationships as is commonly stated in the literature.
title Towards the Characterization of Representations Learned via Capsule-based Network Architectures
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2305.05349