{"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/max-plus-operators-applied-to-filter","title":"Max-plus Operators Applied to Filter Selection and Model Pruning in Neural Networks","arxiv_id":"1903.08072","date":"2019-03-19","proceeding":null,"authors":["Yunxiang Zhang","Samy Blusseau","Santiago Velasco-Forero","Isabelle Bloch","Jesus Angulo"],"abstract":"Following recent advances in morphological neural networks, we propose to\nstudy in more depth how Max-plus operators can be exploited to define\nmorphological units and how they behave when incorporated in layers of\nconventional neural networks. Besides showing that they can be easily\nimplemented with modern machine learning frameworks , we confirm and extend the\nobservation that a Max-plus layer can be used to select important filters and\nreduce redundancy in its previous layer, without incurring performance loss.\nExperimental results demonstrate that the filter selection strategy enabled by\na Max-plus is highly efficient and robust, through which we successfully\nperformed model pruning on different neural network architectures. We also\npoint out that there is a close connection between Maxout networks and our\npruned Max-plus networks by comparing their respective characteristics. The\ncode for reproducing our experiments is available online.","url_abs":"http://arxiv.org/abs/1903.08072v2","url_pdf":"http://arxiv.org/pdf/1903.08072v2.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":"max-plus-operators-applied-to-filter","repo_url":"https://github.com/samyblusseau/maxplusblock","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[],"methods":[{"method_slug":"maxout","method_name":"Maxout"},{"method_slug":"pruning","method_name":"Pruning"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}