{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:LCRCIGGNA72EFQINONZSORLH5U","short_pith_number":"pith:LCRCIGGN","schema_version":"1.0","canonical_sha256":"58a22418cd07f442c10d7373274567ed126f5a4114bd33d13291084400357cfb","source":{"kind":"arxiv","id":"1910.04281","version":2},"attestation_state":"computed","paper":{"title":"Integrating Behavior Cloning and Reinforcement Learning for Improved Performance in Dense and Sparse Reward Environments","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Gregory M. Gremillion, John Valasek, Nicholas R. Waytowich, Vernon J. Lawhern, Vinicius G. Goecks","submitted_at":"2019-10-09T22:32:23Z","abstract_excerpt":"This paper investigates how to efficiently transition and update policies, trained initially with demonstrations, using off-policy actor-critic reinforcement learning. It is well-known that techniques based on Learning from Demonstrations, for example behavior cloning, can lead to proficient policies given limited data. However, it is currently unclear how to efficiently update that policy using reinforcement learning as these approaches are inherently optimizing different objective functions. Previous works have used loss functions, which combine behavior cloning losses with reinforcement lea"},"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":"1910.04281","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-09T22:32:23Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"cf479e6ad0acdaa75a8baa953fd583089d7ec19a636df37d631c2a151e6bbc6a","abstract_canon_sha256":"fbe4e5c9c6aef117133aa105db0ae780439a48df020bcd5487c8883760a9bf56"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:52:36.490970Z","signature_b64":"DFQnLw9GxRIfMtsp8Q+AWsfmE1yDU+0GNbZODsokWxmshxxi0kDy0W5uO/RjeNvB3Sy7RnxFdtIMYohNGUsZAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"58a22418cd07f442c10d7373274567ed126f5a4114bd33d13291084400357cfb","last_reissued_at":"2026-07-05T00:52:36.490637Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:52:36.490637Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Integrating Behavior Cloning and Reinforcement Learning for Improved Performance in Dense and Sparse Reward Environments","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Gregory M. Gremillion, John Valasek, Nicholas R. Waytowich, Vernon J. Lawhern, Vinicius G. Goecks","submitted_at":"2019-10-09T22:32:23Z","abstract_excerpt":"This paper investigates how to efficiently transition and update policies, trained initially with demonstrations, using off-policy actor-critic reinforcement learning. It is well-known that techniques based on Learning from Demonstrations, for example behavior cloning, can lead to proficient policies given limited data. However, it is currently unclear how to efficiently update that policy using reinforcement learning as these approaches are inherently optimizing different objective functions. Previous works have used loss functions, which combine behavior cloning losses with reinforcement lea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.04281","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/1910.04281/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":"1910.04281","created_at":"2026-07-05T00:52:36.490691+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.04281v2","created_at":"2026-07-05T00:52:36.490691+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.04281","created_at":"2026-07-05T00:52:36.490691+00:00"},{"alias_kind":"pith_short_12","alias_value":"LCRCIGGNA72E","created_at":"2026-07-05T00:52:36.490691+00:00"},{"alias_kind":"pith_short_16","alias_value":"LCRCIGGNA72EFQIN","created_at":"2026-07-05T00:52:36.490691+00:00"},{"alias_kind":"pith_short_8","alias_value":"LCRCIGGN","created_at":"2026-07-05T00:52:36.490691+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19656","citing_title":"DF-ExpEnse: Diffusion Filtered Exploration for Sample Efficient Finetuning","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11743","citing_title":"TacCoRL: Integrating Tactile Feedback into VLA via Simulation","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15855","citing_title":"Do Less, Achieve More: Do We Need Every-Step Optimization for RL Fine-tuning of Diffusion Models?","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LCRCIGGNA72EFQINONZSORLH5U","json":"https://pith.science/pith/LCRCIGGNA72EFQINONZSORLH5U.json","graph_json":"https://pith.science/api/pith-number/LCRCIGGNA72EFQINONZSORLH5U/graph.json","events_json":"https://pith.science/api/pith-number/LCRCIGGNA72EFQINONZSORLH5U/events.json","paper":"https://pith.science/paper/LCRCIGGN"},"agent_actions":{"view_html":"https://pith.science/pith/LCRCIGGNA72EFQINONZSORLH5U","download_json":"https://pith.science/pith/LCRCIGGNA72EFQINONZSORLH5U.json","view_paper":"https://pith.science/paper/LCRCIGGN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.04281&json=true","fetch_graph":"https://pith.science/api/pith-number/LCRCIGGNA72EFQINONZSORLH5U/graph.json","fetch_events":"https://pith.science/api/pith-number/LCRCIGGNA72EFQINONZSORLH5U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LCRCIGGNA72EFQINONZSORLH5U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LCRCIGGNA72EFQINONZSORLH5U/action/storage_attestation","attest_author":"https://pith.science/pith/LCRCIGGNA72EFQINONZSORLH5U/action/author_attestation","sign_citation":"https://pith.science/pith/LCRCIGGNA72EFQINONZSORLH5U/action/citation_signature","submit_replication":"https://pith.science/pith/LCRCIGGNA72EFQINONZSORLH5U/action/replication_record"}},"created_at":"2026-07-05T00:52:36.490691+00:00","updated_at":"2026-07-05T00:52:36.490691+00:00"}