{"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/translating-neuralese","title":"Translating Neuralese","arxiv_id":"1704.06960","date":"2017-04-23","proceeding":"ACL 2017 7","authors":["Jacob Andreas","Anca Dragan","Dan Klein"],"abstract":"Several approaches have recently been proposed for learning decentralized\ndeep multiagent policies that coordinate via a differentiable communication\nchannel. While these policies are effective for many tasks, interpretation of\ntheir induced communication strategies has remained a challenge. Here we\npropose to interpret agents' messages by translating them. Unlike in typical\nmachine translation problems, we have no parallel data to learn from. Instead\nwe develop a translation model based on the insight that agent messages and\nnatural language strings mean the same thing if they induce the same belief\nabout the world in a listener. We present theoretical guarantees and empirical\nevidence that our approach preserves both the semantics and pragmatics of\nmessages by ensuring that players communicating through a translation layer do\nnot suffer a substantial loss in reward relative to players with a common\nlanguage.","url_abs":"http://arxiv.org/abs/1704.06960v5","url_pdf":"http://arxiv.org/pdf/1704.06960v5.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":"translating-neuralese","repo_url":"https://github.com/jacobandreas/neuralese","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.06960","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}