{"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/end-to-end-differentiable-proving","title":"End-to-End Differentiable Proving","arxiv_id":"1705.11040","date":"2017-05-31","proceeding":"NeurIPS 2017 12","authors":["Tim Rocktäschel","Sebastian Riedel"],"abstract":"We introduce neural networks for end-to-end differentiable proving of queries\nto knowledge bases by operating on dense vector representations of symbols.\nThese neural networks are constructed recursively by taking inspiration from\nthe backward chaining algorithm as used in Prolog. Specifically, we replace\nsymbolic unification with a differentiable computation on vector\nrepresentations of symbols using a radial basis function kernel, thereby\ncombining symbolic reasoning with learning subsymbolic vector representations.\nBy using gradient descent, the resulting neural network can be trained to infer\nfacts from a given incomplete knowledge base. It learns to (i) place\nrepresentations of similar symbols in close proximity in a vector space, (ii)\nmake use of such similarities to prove queries, (iii) induce logical rules, and\n(iv) use provided and induced logical rules for multi-hop reasoning. We\ndemonstrate that this architecture outperforms ComplEx, a state-of-the-art\nneural link prediction model, on three out of four benchmark knowledge bases\nwhile at the same time inducing interpretable function-free first-order logic\nrules.","url_abs":"http://arxiv.org/abs/1705.11040v2","url_pdf":"http://arxiv.org/pdf/1705.11040v2.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":"end-to-end-differentiable-proving","repo_url":"https://github.com/uclmr/ntp","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"end-to-end-differentiable-proving","repo_url":"https://github.com/Michiel29/ntp-release","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"end-to-end-differentiable-proving","repo_url":"https://github.com/chanind/tensor-theorem-prover","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"link-prediction","task_name":"Link Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1705.11040","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}