{"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/arguing-machines-human-supervision-of-black","title":"Arguing Machines: Human Supervision of Black Box AI Systems That Make Life-Critical Decisions","arxiv_id":"1710.04459","date":"2017-10-12","proceeding":null,"authors":["Lex Fridman","Li Ding","Benedikt Jenik","Bryan Reimer"],"abstract":"We consider the paradigm of a black box AI system that makes life-critical\ndecisions. We propose an \"arguing machines\" framework that pairs the primary AI\nsystem with a secondary one that is independently trained to perform the same\ntask. We show that disagreement between the two systems, without any knowledge\nof underlying system design or operation, is sufficient to arbitrarily improve\nthe accuracy of the overall decision pipeline given human supervision over\ndisagreements. We demonstrate this system in two applications: (1) an\nillustrative example of image classification and (2) on large-scale real-world\nsemi-autonomous driving data. For the first application, we apply this\nframework to image classification achieving a reduction from 8.0% to 2.8% top-5\nerror on ImageNet. For the second application, we apply this framework to Tesla\nAutopilot and demonstrate the ability to predict 90.4% of system disengagements\nthat were labeled by human annotators as challenging and needing human\nsupervision.","url_abs":"http://arxiv.org/abs/1710.04459v2","url_pdf":"http://arxiv.org/pdf/1710.04459v2.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":"arguing-machines-human-supervision-of-black","repo_url":"https://github.com/scope-lab-vu/deep-nn-car","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"autonomous-driving","task_name":"Autonomous Driving"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}