Capsule Networks are an alternative to the conventional CNN structure for object recognition. They replace max pooling with a dynamic routing of capsule activation. The goal is to better exploit the spatial relationships of the learned features, not only to increase recognition performance, but also improve generalization capability and sample-efficiency. Recently, two algorithms for dynamic routing of capsules have been proposed. Although they received a lot of interest and they are from the same group, an experimental comparison of both is still missing. In this work we compare these two routing algorithms and
provide experimental results on data efficiency and generalization to increased input images. Although the experiments are limited to variants of the MNIST dataset, they indicate that the approach of Sabour et al. (2017) is better at learning from few training samples and the EM routing of Hinton et al. (2018) is better at generalizing to changed image sizes.
| Identifer | oai:union.ndltd.org:DRESDEN/oai:qucosa:de:qucosa:32763 |
| Date | 22 January 2019 |
| Creators | Schlegel, Kenny, Neubert, Peer, Protzel, Peter |
| Contributors | Technische Universität Chemnitz |
| Source Sets | Hochschulschriftenserver (HSSS) der SLUB Dresden |
| Language | English |
| Detected Language | English |
| Type | info:eu-repo/semantics/acceptedVersion, doc-type:conferenceObject, info:eu-repo/semantics/conferenceObject, doc-type:Text |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0016 seconds