{"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/multi-agent-imitation-learning-for-driving","title":"Multi-Agent Imitation Learning for Driving Simulation","arxiv_id":"1803.01044","date":"2018-03-02","proceeding":null,"authors":["Raunak P. Bhattacharyya","Derek J. Phillips","Blake Wulfe","Jeremy Morton","Alex Kuefler","Mykel J. Kochenderfer"],"abstract":"Simulation is an appealing option for validating the safety of autonomous\nvehicles. Generative Adversarial Imitation Learning (GAIL) has recently been\nshown to learn representative human driver models. These human driver models\nwere learned through training in single-agent environments, but they have\ndifficulty in generalizing to multi-agent driving scenarios. We argue these\ndifficulties arise because observations at training and test time are sampled\nfrom different distributions. This difference makes such models unsuitable for\nthe simulation of driving scenes, where multiple agents must interact\nrealistically over long time horizons. We extend GAIL to address these\nshortcomings through a parameter-sharing approach grounded in curriculum\nlearning. Compared with single-agent GAIL policies, policies generated by our\nPS-GAIL method prove superior at interacting stably in a multi-agent setting\nand capturing the emergent behavior of human drivers.","url_abs":"http://arxiv.org/abs/1803.01044v1","url_pdf":"http://arxiv.org/pdf/1803.01044v1.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":"multi-agent-imitation-learning-for-driving","repo_url":"https://github.com/sisl/ngsim_env","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"autonomous-vehicles","task_name":"Autonomous Vehicles"},{"task_slug":"imitation-learning","task_name":"Imitation Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1803.01044","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}