{"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/a-learnable-scatternet-locally-invariant","title":"A Learnable ScatterNet: Locally Invariant Convolutional Layers","arxiv_id":"1903.03137","date":"2019-03-07","proceeding":null,"authors":["Fergal Cotter","Nick Kingsbury"],"abstract":"In this paper we explore tying together the ideas from Scattering Transforms\nand Convolutional Neural Networks (CNN) for Image Analysis by proposing a\nlearnable ScatterNet. Previous attempts at tying them together in hybrid\nnetworks have tended to keep the two parts separate, with the ScatterNet\nforming a fixed front end and a CNN forming a learned backend. We instead look\nat adding learning between scattering orders, as well as adding learned layers\nbefore the ScatterNet. We do this by breaking down the scattering orders into\nsingle convolutional-like layers we call 'locally invariant' layers, and adding\na learned mixing term to this layer. Our experiments show that these locally\ninvariant layers can improve accuracy when added to either a CNN or a\nScatterNet. We also discover some surprising results in that the ScatterNet may\nbe best positioned after one or more layers of learning rather than at the\nfront of a neural network.","url_abs":"http://arxiv.org/abs/1903.03137v1","url_pdf":"http://arxiv.org/pdf/1903.03137v1.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":"a-learnable-scatternet-locally-invariant","repo_url":"https://github.com/fbcotter/scatnet_learn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.03137","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}