{"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/a-critical-investigation-of-deep","title":"A Critical Investigation of Deep Reinforcement Learning for Navigation","arxiv_id":"1802.02274","date":"2018-02-07","proceeding":null,"authors":["Vikas Dhiman","Shurjo Banerjee","Brent Griffin","Jeffrey M. Siskind","Jason J. Corso"],"abstract":"The navigation problem is classically approached in two steps: an exploration\nstep, where map-information about the environment is gathered; and an\nexploitation step, where this information is used to navigate efficiently. Deep\nreinforcement learning (DRL) algorithms, alternatively, approach the problem of\nnavigation in an end-to-end fashion. Inspired by the classical approach, we ask\nwhether DRL algorithms are able to inherently explore, gather and exploit\nmap-information over the course of navigation. We build upon Mirowski et al.\n[2017] work and introduce a systematic suite of experiments that vary three\nparameters: the agent's starting location, the agent's target location, and the\nmaze structure. We choose evaluation metrics that explicitly measure the\nalgorithm's ability to gather and exploit map-information. Our experiments show\nthat when trained and tested on the same maps, the algorithm successfully\ngathers and exploits map-information. However, when trained and tested on\ndifferent sets of maps, the algorithm fails to transfer the ability to gather\nand exploit map-information to unseen maps. Furthermore, we find that when the\ngoal location is randomized and the map is kept static, the algorithm is able\nto gather and exploit map-information but the exploitation is far from optimal.\nWe open-source our experimental suite in the hopes that it serves as a\nframework for the comparison of future algorithms and leads to the discovery of\nrobust alternatives to classical navigation methods.","url_abs":"http://arxiv.org/abs/1802.02274v2","url_pdf":"http://arxiv.org/pdf/1802.02274v2.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":"a-critical-investigation-of-deep","repo_url":"https://github.com/umrobotslang/does-drl-learn-to-navigate","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"navigate","task_name":"Navigate"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}