{"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/modularity-as-a-means-for-complexity","title":"Modularity as a Means for Complexity Management in Neural Networks Learning","arxiv_id":"1902.09240","date":"2019-02-25","proceeding":null,"authors":["David Castillo-Bolado","Cayetano Guerra-Artal","Mario Hernandez-Tejera"],"abstract":"Training a Neural Network (NN) with lots of parameters or intricate\narchitectures creates undesired phenomena that complicate the optimization\nprocess. To address this issue we propose a first modular approach to NN\ndesign, wherein the NN is decomposed into a control module and several\nfunctional modules, implementing primitive operations. We illustrate the\nmodular concept by comparing performances between a monolithic and a modular NN\non a list sorting problem and show the benefits in terms of training speed,\ntraining stability and maintainability. We also discuss some questions that\narise in modular NNs.","url_abs":"http://arxiv.org/abs/1902.09240v1","url_pdf":"http://arxiv.org/pdf/1902.09240v1.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":"modularity-as-a-means-for-complexity","repo_url":"https://gitlab.com/dcasbol/nn-modularization","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"management","task_name":"Management"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}