{"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/hierarchical-density-order-embeddings","title":"Hierarchical Density Order Embeddings","arxiv_id":"1804.09843","date":"2018-04-26","proceeding":"ICLR 2018 1","authors":["Ben Athiwaratkun","Andrew Gordon Wilson"],"abstract":"By representing words with probability densities rather than point vectors,\nprobabilistic word embeddings can capture rich and interpretable semantic\ninformation and uncertainty. The uncertainty information can be particularly\nmeaningful in capturing entailment relationships -- whereby general words such\nas \"entity\" correspond to broad distributions that encompass more specific\nwords such as \"animal\" or \"instrument\". We introduce density order embeddings,\nwhich learn hierarchical representations through encapsulation of probability\ndensities. In particular, we propose simple yet effective loss functions and\ndistance metrics, as well as graph-based schemes to select negative samples to\nbetter learn hierarchical density representations. Our approach provides\nstate-of-the-art performance on the WordNet hypernym relationship prediction\ntask and the challenging HyperLex lexical entailment dataset -- while retaining\na rich and interpretable density representation.","url_abs":"http://arxiv.org/abs/1804.09843v1","url_pdf":"http://arxiv.org/pdf/1804.09843v1.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":"hierarchical-density-order-embeddings","repo_url":"https://github.com/benathi/density-order-emb","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"hierarchical-density-order-embeddings","repo_url":"https://github.com/jvdbogae/artverc","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"lexical-entailment","task_name":"Lexical Entailment"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.09843","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}