{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/improved-efficient-capsule-network-for","title":"Improved efficient capsule network for Kuzushiji-MNIST benchmark dataset classification","arxiv_id":null,"date":"2023-12-15","proceeding":"Bulletin of the Polish Academy of Sciences Technical Sciences 2023 12","authors":["Michał Bukowski","Izabella Antoniuk","Jarosław Kurek"],"abstract":"In this paper, we present an improved efficient capsule network (CN) model for the classification of the Kuzushiji-MNIST and\r\nKuzushiji-49 benchmark datasets. CNs are a promising approach in the field of deep learning, offering advantages such as robustness, better\r\ngeneralization, and a simpler network structure compared to traditional convolutional neural networks (CNNs). Proposed model, based on the\r\nEfficient CapsNet architecture, incorporates the self-attention routing mechanism, resulting in improved efficiency and reduced parameter count.\r\nThe experiments conducted on the Kuzushiji-MNIST and Kuzushiji-49 datasets demonstrate that the model achieves competitive performance,\r\nranking within the top ten solutions for both benchmarks. Despite using significantly fewer parameters compared to higher-rated competitors,\r\npresented model achieves comparable accuracy, with overall differences of only 0.91% and 1.97% for the Kuzushiji-MNIST and Kuzushiji-\r\n49 datasets, respectively. Furthermore, the training time required to achieve these results is substantially reduced, enabling training on non-\r\nspecialized workstations. The proposed novelties of capsule architecture, including the integration of the self-attention mechanism and the\r\nefficient network structure, contribute to the improved efficiency and performance of presented model. These findings highlight the potential of\r\nCNs as a more efficient and effective approach for character classification tasks, with broader applications in various domains.","url_abs":"https://journals.pan.pl/dlibra/publication/147338/edition/128840/content","url_pdf":"https://journals.pan.pl/dlibra/publication/147338/edition/128840/content","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"improved-efficient-capsule-network-for","repo_url":"https://github.com/bukson/kmnist-efcaps","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"}],"methods":[{"method_slug":"fixcaps","method_name":"CapsNet"},{"method_slug":"fixcaps","method_name":"Capsule Network"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-kuzushiji-mnist","task":"Image Classification","dataset":"Kuzushiji-MNIST","model":"Efficient Capsnet","rank_in_archive_order":18,"of":26,"metrics":{"Accuracy":"98.43"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}