{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:VWWHPB3A7LDFLZSEEFRFGEGNKQ","short_pith_number":"pith:VWWHPB3A","schema_version":"1.0","canonical_sha256":"adac778760fac655e64421625310cd543fe4e309fa54574d5dc0b75f4cc8ad54","source":{"kind":"arxiv","id":"1905.11979","version":2},"attestation_state":"computed","paper":{"title":"Causal Confusion in Imitation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dinesh Jayaraman, Pim de Haan, Sergey Levine","submitted_at":"2019-05-28T17:56:19Z","abstract_excerpt":"Behavioral cloning reduces policy learning to supervised learning by training a discriminative model to predict expert actions given observations. Such discriminative models are non-causal: the training procedure is unaware of the causal structure of the interaction between the expert and the environment. We point out that ignoring causality is particularly damaging because of the distributional shift in imitation learning. In particular, it leads to a counter-intuitive \"causal misidentification\" phenomenon: access to more information can yield worse performance. We investigate how this proble"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1905.11979","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-05-28T17:56:19Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"e591589acc60cda46ec6dfa4b46f6eef52ee3a144cfb70e47baab13b2391c907","abstract_canon_sha256":"4025d313cdd913f880aa500bc343722abc32fe9883832420b3fdf68321cda796"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:16:34.729432Z","signature_b64":"cqAHjg9CCe8llBnPyBxYYllT/BafvcRMW4YRSZ4FLAnnYD2JK4b9AhkV2jbDegm8brMiqp/FfhL6N9aNsgxAAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"adac778760fac655e64421625310cd543fe4e309fa54574d5dc0b75f4cc8ad54","last_reissued_at":"2026-07-05T00:16:34.728999Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:16:34.728999Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Causal Confusion in Imitation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dinesh Jayaraman, Pim de Haan, Sergey Levine","submitted_at":"2019-05-28T17:56:19Z","abstract_excerpt":"Behavioral cloning reduces policy learning to supervised learning by training a discriminative model to predict expert actions given observations. Such discriminative models are non-causal: the training procedure is unaware of the causal structure of the interaction between the expert and the environment. We point out that ignoring causality is particularly damaging because of the distributional shift in imitation learning. In particular, it leads to a counter-intuitive \"causal misidentification\" phenomenon: access to more information can yield worse performance. We investigate how this proble"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.11979","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1905.11979/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"1905.11979","created_at":"2026-07-05T00:16:34.729054+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.11979v2","created_at":"2026-07-05T00:16:34.729054+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.11979","created_at":"2026-07-05T00:16:34.729054+00:00"},{"alias_kind":"pith_short_12","alias_value":"VWWHPB3A7LDF","created_at":"2026-07-05T00:16:34.729054+00:00"},{"alias_kind":"pith_short_16","alias_value":"VWWHPB3A7LDFLZSE","created_at":"2026-07-05T00:16:34.729054+00:00"},{"alias_kind":"pith_short_8","alias_value":"VWWHPB3A","created_at":"2026-07-05T00:16:34.729054+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.23565","citing_title":"Understanding Goal Generalisation in Sequential Reinforcement Learning","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14598","citing_title":"DSSP: Diffusion State Space Policy with Full-History Encoding","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VWWHPB3A7LDFLZSEEFRFGEGNKQ","json":"https://pith.science/pith/VWWHPB3A7LDFLZSEEFRFGEGNKQ.json","graph_json":"https://pith.science/api/pith-number/VWWHPB3A7LDFLZSEEFRFGEGNKQ/graph.json","events_json":"https://pith.science/api/pith-number/VWWHPB3A7LDFLZSEEFRFGEGNKQ/events.json","paper":"https://pith.science/paper/VWWHPB3A"},"agent_actions":{"view_html":"https://pith.science/pith/VWWHPB3A7LDFLZSEEFRFGEGNKQ","download_json":"https://pith.science/pith/VWWHPB3A7LDFLZSEEFRFGEGNKQ.json","view_paper":"https://pith.science/paper/VWWHPB3A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.11979&json=true","fetch_graph":"https://pith.science/api/pith-number/VWWHPB3A7LDFLZSEEFRFGEGNKQ/graph.json","fetch_events":"https://pith.science/api/pith-number/VWWHPB3A7LDFLZSEEFRFGEGNKQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VWWHPB3A7LDFLZSEEFRFGEGNKQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VWWHPB3A7LDFLZSEEFRFGEGNKQ/action/storage_attestation","attest_author":"https://pith.science/pith/VWWHPB3A7LDFLZSEEFRFGEGNKQ/action/author_attestation","sign_citation":"https://pith.science/pith/VWWHPB3A7LDFLZSEEFRFGEGNKQ/action/citation_signature","submit_replication":"https://pith.science/pith/VWWHPB3A7LDFLZSEEFRFGEGNKQ/action/replication_record"}},"created_at":"2026-07-05T00:16:34.729054+00:00","updated_at":"2026-07-05T00:16:34.729054+00:00"}