{"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/automatic-inference-of-cross-modal-connection","title":"Automatic Inference of Cross-modal Connection Topologies for X-CNNs","arxiv_id":"1805.00987","date":"2018-05-02","proceeding":null,"authors":["Laurynas Karazija","Petar Veličković","Pietro Liò"],"abstract":"This paper introduces a way to learn cross-modal convolutional neural network\n(X-CNN) architectures from a base convolutional network (CNN) and the training\ndata to reduce the design cost and enable applying cross-modal networks in\nsparse data environments. Two approaches for building X-CNNs are presented. The\nbase approach learns the topology in a data-driven manner, by using\nmeasurements performed on the base CNN and supplied data. The iterative\napproach performs further optimisation of the topology through a combined\nlearning procedure, simultaneously learning the topology and training the\nnetwork. The approaches were evaluated agains examples of hand-designed X-CNNs\nand their base variants, showing superior performance and, in some cases,\ngaining an additional 9% of accuracy. From further considerations, we conclude\nthat the presented methodology takes less time than any manual approach would,\nwhilst also significantly reducing the design complexity. The application of\nthe methods is fully automated and implemented in Xsertion library.","url_abs":"http://arxiv.org/abs/1805.00987v1","url_pdf":"http://arxiv.org/pdf/1805.00987v1.pdf","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":"automatic-inference-of-cross-modal-connection","repo_url":"https://github.com/karazijal/xsertion","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}