A4 Article in conference proceedings
Causality-Aware Convolutional Neural Networks for Advanced Image Classification and Generation (2023)


Terziyan, 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.), 4th International Conference on Industry 4.0 and Smart Manufacturing (pp. 495-506). Elsevier. Procedia Computer Science, 217. https://doi.org/10.1016/j.procs.2022.12.245


JYU authors or editors


Publication details

All authors or editorsTerziyan, Vagan; Vitko, Oleksandra

Parent publication4th International Conference on Industry 4.0 and Smart Manufacturing

Parent publication editorsLongo, Francesco; Affenzeller, Michael; Padovano, Antonio; Weiming, Shen

Conference:

  • International Conference on Industry 4.0 and Smart Manufacturing

Place and date of conferenceLinz, Austria2.-4.11.2022

Journal or seriesProcedia Computer Science

eISSN1877-0509

Publication year2023

Publication date13/01/2023

Number in series217

Pages range495-506

Number of pages in the book1954

PublisherElsevier

Publication countryNetherlands

Publication languageEnglish

DOIhttps://doi.org/10.1016/j.procs.2022.12.245

Publication open accessOpenly available

Publication channel open accessOpen Access channel

Publication is parallel published (JYX)https://jyx.jyu.fi/handle/123456789/85106


Abstract

Smart 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.pptx


Keywordsmanufacturing engineeringcomputer visionmachine learningdeep learningneural networks (information technology)classificationcausalityinference

Free keywordscausal discovery; causal inference; image processing; Convolutional Neural Network; Generative Adversarial Network


Contributing organizations


Ministry reportingYes

VIRTA submission year2023

JUFO rating1


Last updated on 2024-12-10 at 18:00