Show simple item record

dc.contributor.authorTerziyan, Vagan
dc.contributor.authorVitko, Oleksandra
dc.contributor.editorLongo, Francesco
dc.contributor.editorAffenzeller, Michael
dc.contributor.editorPadovano, Antonio
dc.contributor.editorWeiming, Shen
dc.date.accessioned2023-01-19T12:31:29Z
dc.date.available2023-01-19T12:31:29Z
dc.date.issued2023
dc.identifier.citationTerziyan, V., & Vitko, O. (2023). Causality-Aware Convolutional Neural Networks for Advanced Image Classification and Generation. In F. Longo, M. Affenzeller, A. Padovano, & S. Weiming (Eds.), <i>4th International Conference on Industry 4.0 and Smart Manufacturing</i> (pp. 495-506). Elsevier. Procedia Computer Science, 217. <a href="https://doi.org/10.1016/j.procs.2022.12.245" target="_blank">https://doi.org/10.1016/j.procs.2022.12.245</a>
dc.identifier.otherCONVID_172578730
dc.identifier.urihttps://jyx.jyu.fi/handle/123456789/85106
dc.description.abstractSmart manufacturing uses emerging deep learning models, and particularly Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs), for different industrial diagnostics tasks, e.g., classification, detection, recognition, prediction, synthetic data generation, security, etc., on the basis of image data. In spite of being efficient for these objectives, the majority of current deep learning models lack interpretability and explainability. They can discover features hidden within input data together with their mutual co-occurrence. However, they are weak at discovering and making explicit hidden causalities between the features, which could be the reason behind the particular diagnoses. In this paper, we suggest Causality-Aware CNNs (CA-CNNs) and Causality-Aware GANs (CA-GANs) to address the issue of learning hidden causalities within images. The core architecture includes an additional layer of neurons (after the last convolution-pooling and just before the dense layers), which learns pairwise conditional probabilities (aka causality estimates) for the features. Computations for these neurons are driven by the adaptive Lehmer mean function. Learned causalities are merged with the features during flattening and (via fully connected layers) influence the classification outcomes. Such causality estimates can be done for the mixed inputs where images are combined with other data. We argue that CA-CNNs not only improve the classification performance of normal CNNs but also open additional opportunities for the explainability of the models’ outcomes. We consider as an additional advantage for CA-CNNs (if used as a discriminator within CA-GANs) the possibility to generate realistically looking images with respect to the causalities. See presentation slides: https://ai.it.jyu.fi/ISM-2022-Causality.pptxen
dc.format.extent1954
dc.format.mimetypeapplication/pdf
dc.language.isoeng
dc.publisherElsevier
dc.relation.ispartof4th International Conference on Industry 4.0 and Smart Manufacturing
dc.relation.ispartofseriesProcedia Computer Science
dc.rightsCC BY-NC-ND 4.0
dc.subject.othercausal discovery
dc.subject.othercausal inference
dc.subject.otherimage processing
dc.subject.otherConvolutional Neural Network
dc.subject.otherGenerative Adversarial Network
dc.titleCausality-Aware Convolutional Neural Networks for Advanced Image Classification and Generation
dc.typeconferenceObject
dc.identifier.urnURN:NBN:fi:jyu-202301191406
dc.contributor.laitosInformaatioteknologian tiedekuntafi
dc.contributor.laitosFaculty of Information Technologyen
dc.contributor.oppiaineCollective Intelligencefi
dc.contributor.oppiaineTekniikkafi
dc.contributor.oppiaineCollective Intelligenceen
dc.contributor.oppiaineEngineeringen
dc.type.urihttp://purl.org/eprint/type/ConferencePaper
dc.type.coarhttp://purl.org/coar/resource_type/c_5794
dc.description.reviewstatuspeerReviewed
dc.format.pagerange495-506
dc.relation.issn1877-0509
dc.type.versionpublishedVersion
dc.rights.copyright© 2022 The Authors. Published by Elsevier B.V.
dc.rights.accesslevelopenAccessfi
dc.relation.conferenceInternational Conference on Industry 4.0 and Smart Manufacturing
dc.subject.ysoneuroverkot
dc.subject.ysosyväoppiminen
dc.subject.ysoluokitus (toiminta)
dc.subject.ysovalmistustekniikka
dc.subject.ysokoneoppiminen
dc.subject.ysokonenäkö
dc.subject.ysokausaliteetti
dc.subject.ysopäättely
dc.format.contentfulltext
jyx.subject.urihttp://www.yso.fi/onto/yso/p7292
jyx.subject.urihttp://www.yso.fi/onto/yso/p39324
jyx.subject.urihttp://www.yso.fi/onto/yso/p12668
jyx.subject.urihttp://www.yso.fi/onto/yso/p22012
jyx.subject.urihttp://www.yso.fi/onto/yso/p21846
jyx.subject.urihttp://www.yso.fi/onto/yso/p2618
jyx.subject.urihttp://www.yso.fi/onto/yso/p333
jyx.subject.urihttp://www.yso.fi/onto/yso/p5902
dc.rights.urlhttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.relation.doi10.1016/j.procs.2022.12.245
dc.type.okmA4


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

CC BY-NC-ND 4.0
Except where otherwise noted, this item's license is described as CC BY-NC-ND 4.0