{"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/behavioral-cloning-from-observation","title":"Behavioral Cloning from Observation","arxiv_id":"1805.01954","date":"2018-05-04","proceeding":null,"authors":["Faraz Torabi","Garrett Warnell","Peter Stone"],"abstract":"Humans often learn how to perform tasks via imitation: they observe others\nperform a task, and then very quickly infer the appropriate actions to take\nbased on their observations. While extending this paradigm to autonomous agents\nis a well-studied problem in general, there are two particular aspects that\nhave largely been overlooked: (1) that the learning is done from observation\nonly (i.e., without explicit action information), and (2) that the learning is\ntypically done very quickly. In this work, we propose a two-phase, autonomous\nimitation learning technique called behavioral cloning from observation (BCO),\nthat aims to provide improved performance with respect to both of these\naspects. First, we allow the agent to acquire experience in a self-supervised\nfashion. This experience is used to develop a model which is then utilized to\nlearn a particular task by observing an expert perform that task without the\nknowledge of the specific actions taken. We experimentally compare BCO to\nimitation learning methods, including the state-of-the-art, generative\nadversarial imitation learning (GAIL) technique, and we show comparable task\nperformance in several different simulation domains while exhibiting increased\nlearning speed after expert trajectories become available.","url_abs":"http://arxiv.org/abs/1805.01954v2","url_pdf":"http://arxiv.org/pdf/1805.01954v2.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":"behavioral-cloning-from-observation","repo_url":"https://github.com/NathanGavenski/IUPE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"behavioral-cloning-from-observation","repo_url":"https://github.com/baichenjia/utds","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"paper_slug":"behavioral-cloning-from-observation","repo_url":"https://github.com/greerviau/GlorifiedCruiseControl","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"behavioral-cloning-from-observation","repo_url":"https://github.com/jerrylin1121/BCO","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"behavioral-cloning-from-observation","repo_url":"https://github.com/montaserFath/BCO","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"behavioral-cloning-from-observation","repo_url":"https://github.com/opendilab/DI-engine","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"imitation-learning","task_name":"Imitation Learning"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.01954","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1805.01954"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/NathanGavenski/IUPE","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jerrylin1121/BCO","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/greerviau/GlorifiedCruiseControl","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/baichenjia/utds","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/montaserFath/BCO","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/opendilab/DI-engine","reach":null}],"summary":{"ran_draft_wrong":2,"ran_honours":2,"unverified":3},"by_repo_kind":{"listed":{"samples":7,"ran":4,"repositories":2}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"65faa99ef0c486da","entry":"train","repo":"NathanGavenski/IUPE","repo_kind":"listed","path":"models/policy.py","file_url":"https://github.com/NathanGavenski/IUPE/blob/HEAD/models/policy.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"65faa99ef0c486da"}},{"code_sha256_prefix":"6cb5bfd9f6b92a48","entry":"train","repo":"NathanGavenski/IUPE","repo_kind":"listed","path":"models/idm.py","file_url":"https://github.com/NathanGavenski/IUPE/blob/HEAD/models/idm.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6cb5bfd9f6b92a48"}},{"code_sha256_prefix":"5c68d61a253a334d","entry":"validation","repo":"NathanGavenski/IUPE","repo_kind":"listed","path":"models/policy.py","file_url":"https://github.com/NathanGavenski/IUPE/blob/HEAD/models/policy.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5c68d61a253a334d"}},{"code_sha256_prefix":"a90673e1d18b15a3","entry":"validation","repo":"NathanGavenski/IUPE","repo_kind":"listed","path":"models/idm.py","file_url":"https://github.com/NathanGavenski/IUPE/blob/HEAD/models/idm.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a90673e1d18b15a3"}},{"code_sha256_prefix":"10e8814dfd1ac64c","entry":"average_gradients","repo":"greerviau/GlorifiedCruiseControl","repo_kind":"listed","path":"SCNN_lanenet/train_lanenet.py","file_url":"https://github.com/greerviau/GlorifiedCruiseControl/blob/HEAD/SCNN_lanenet/train_lanenet.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"10e8814dfd1ac64c"}},{"code_sha256_prefix":"a4a0b8d9a39afc8c","entry":"clipper","repo":"greerviau/GlorifiedCruiseControl","repo_kind":"listed","path":"data_collection/clip_videos.py","file_url":"https://github.com/greerviau/GlorifiedCruiseControl/blob/HEAD/data_collection/clip_videos.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a4a0b8d9a39afc8c"}},{"code_sha256_prefix":"569798d0afd5c026","entry":"forward","repo":"greerviau/GlorifiedCruiseControl","repo_kind":"listed","path":"SCNN_lanenet/train_lanenet.py","file_url":"https://github.com/greerviau/GlorifiedCruiseControl/blob/HEAD/SCNN_lanenet/train_lanenet.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"569798d0afd5c026"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}