{"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/deep-learning-long-range-information-in","title":"Deep learning long-range information in undirected graphs with wave networks","arxiv_id":"1810.12153","date":"2018-10-29","proceeding":null,"authors":["Matthew K. Matlock","Arghya Datta","Na Le Dang","Kevin Jiang","S. Joshua Swamidass"],"abstract":"Graph algorithms are key tools in many fields of science and technology. Some\nof these algorithms depend on propagating information between distant nodes in\na graph. Recently, there have been a number of deep learning architectures\nproposed to learn on undirected graphs. However, most of these architectures\naggregate information in the local neighborhood of a node, and therefore they\nmay not be capable of efficiently propagating long-range information. To solve\nthis problem we examine a recently proposed architecture, wave, which\npropagates information back and forth across an undirected graph in waves of\nnonlinear computation. We compare wave to graph convolution, an architecture\nbased on local aggregation, and find that wave learns three different\ngraph-based tasks with greater efficiency and accuracy. These three tasks\ninclude (1) labeling a path connecting two nodes in a graph, (2) solving a maze\npresented as an image, and (3) computing voltages in a circuit. These tasks\nrange from trivial to very difficult, but wave can extrapolate from small\ntraining examples to much larger testing examples. These results show that wave\nmay be able to efficiently solve a wide range of problems that require\nlong-range information propagation across undirected graphs. An implementation\nof the wave network, and example code for the maze problem are included in the\ntflon deep learning toolkit (https://bitbucket.org/mkmatlock/tflon).","url_abs":"http://arxiv.org/abs/1810.12153v1","url_pdf":"http://arxiv.org/pdf/1810.12153v1.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":"deep-learning-long-range-information-in","repo_url":"https://bitbucket.org/mkmatlock/tflon","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","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}