{"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/reduction-of-parameter-redundancy-in-biaffine","title":"Reduction of Parameter Redundancy in Biaffine Classifiers with Symmetric and Circulant Weight Matrices","arxiv_id":"1810.08307","date":"2018-10-18","proceeding":"PACLIC 2018 12","authors":["Tomoki Matsuno","Katsuhiko Hayashi","Takahiro Ishihara","Hitoshi Manabe","Yuji Matsumoto"],"abstract":"Currently, the biaffine classifier has been attracting attention as a method\nto introduce an attention mechanism into the modeling of binary relations. For\ninstance, in the field of dependency parsing, the Deep Biaffine Parser by Dozat\nand Manning has achieved state-of-the-art performance as a graph-based\ndependency parser on the English Penn Treebank and CoNLL 2017 shared task. On\nthe other hand, it is reported that parameter redundancy in the weight matrix\nin biaffine classifiers, which has O(n^2) parameters, results in overfitting (n\nis the number of dimensions). In this paper, we attempted to reduce the\nparameter redundancy by assuming either symmetry or circularity of weight\nmatrices. In our experiments on the CoNLL 2017 shared task dataset, our model\nachieved better or comparable accuracy on most of the treebanks with more than\n16% parameter reduction.","url_abs":"http://arxiv.org/abs/1810.08307v1","url_pdf":"http://arxiv.org/pdf/1810.08307v1.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":"reduction-of-parameter-redundancy-in-biaffine","repo_url":"https://github.com/TomokiMatsuno/PACLIC32","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"dependency-parsing","task_name":"Dependency Parsing"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}