{"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/combining-two-and-three-way-embeddings-models","title":"Combining Two And Three-Way Embeddings Models for Link Prediction in Knowledge Bases","arxiv_id":"1506.00999","date":"2015-06-02","proceeding":null,"authors":["Alberto Garcia-Duran","Antoine Bordes","Nicolas Usunier","Yves GRANDVALET"],"abstract":"This paper tackles the problem of endogenous link prediction for Knowledge\nBase completion. Knowledge Bases can be represented as directed graphs whose\nnodes correspond to entities and edges to relationships. Previous attempts\neither consist of powerful systems with high capacity to model complex\nconnectivity patterns, which unfortunately usually end up overfitting on rare\nrelationships, or in approaches that trade capacity for simplicity in order to\nfairly model all relationships, frequent or not. In this paper, we propose\nTatec a happy medium obtained by complementing a high-capacity model with a\nsimpler one, both pre-trained separately and then combined. We present several\nvariants of this model with different kinds of regularization and combination\nstrategies and show that this approach outperforms existing methods on\ndifferent types of relationships by achieving state-of-the-art results on four\nbenchmarks of the literature.","url_abs":"http://arxiv.org/abs/1506.00999v1","url_pdf":"http://arxiv.org/pdf/1506.00999v1.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":"combining-two-and-three-way-embeddings-models","repo_url":"https://github.com/glorotxa/SME","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"combining-two-and-three-way-embeddings-models","repo_url":"https://github.com/usherwang02/SemanticMatchingEnergy-Theano","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"knowledge-base-completion","task_name":"Knowledge Base Completion"},{"task_slug":"link-prediction","task_name":"Link Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}