{"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/global-weisfeiler-lehman-graph-kernels","title":"Global Weisfeiler-Lehman Graph Kernels","arxiv_id":"1703.02379","date":"2017-03-07","proceeding":null,"authors":["Christopher Morris","Kristian Kersting","Petra Mutzel"],"abstract":"Most state-of-the-art graph kernels only take local graph properties into\naccount, i.e., the kernel is computed with regard to properties of the\nneighborhood of vertices or other small substructures. On the other hand,\nkernels that do take global graph propertiesinto account may not scale well to\nlarge graph databases. Here we propose to start exploring the space between\nlocal and global graph kernels, striking the balance between both worlds.\nSpecifically, we introduce a novel graph kernel based on the $k$-dimensional\nWeisfeiler-Lehman algorithm. Unfortunately, the $k$-dimensional\nWeisfeiler-Lehman algorithm scales exponentially in $k$. Consequently, we\ndevise a stochastic version of the kernel with provable approximation\nguarantees using conditional Rademacher averages. On bounded-degree graphs, it\ncan even be computed in constant time. We support our theoretical results with\nexperiments on several graph classification benchmarks, showing that our\nkernels often outperform the state-of-the-art in terms of classification\naccuracies.","url_abs":"http://arxiv.org/abs/1703.02379v3","url_pdf":"http://arxiv.org/pdf/1703.02379v3.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":"global-weisfeiler-lehman-graph-kernels","repo_url":"https://github.com/chrsmrrs/glocalwl","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"graph-classification","task_name":"Graph Classification"}],"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}