{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:DRSOJL5TL5TVNSJEZPZWQ6M3UP","short_pith_number":"pith:DRSOJL5T","schema_version":"1.0","canonical_sha256":"1c64e4afb35f6756c924cbf368799ba3fbc7a469280c97b76332794873b64b87","source":{"kind":"arxiv","id":"2205.11019","version":1},"attestation_state":"computed","paper":{"title":"Efficient Reinforcement Learning from Demonstration Using Local Ensemble and Reparameterization with Split and Merge of Expert Policies","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Fang Liu, Yu Wang","submitted_at":"2022-05-23T03:36:24Z","abstract_excerpt":"The current work on reinforcement learning (RL) from demonstrations often assumes the demonstrations are samples from an optimal policy, an unrealistic assumption in practice. When demonstrations are generated by sub-optimal policies or have sparse state-action pairs, policy learned from sub-optimal demonstrations may mislead an agent with incorrect or non-local action decisions. We propose a new method called Local Ensemble and Reparameterization with Split and Merge of expert policies (LEARN-SAM) to improve efficiency and make better use of the sub-optimal demonstrations. First, LEARN-SAM em"},"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":"2205.11019","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2022-05-23T03:36:24Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"ae106808fd1537cdb075bb44826bb19cbf2732f731a85a03ea8af3ecf49ffc21","abstract_canon_sha256":"e625cd3080bd9fb3eadf93eb772d106b3d3411fee0989efae101c66e89f01d4e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:25:26.384574Z","signature_b64":"L/KWAOOluvQ+7Ak2DtcvOux1Yr6taFCiN3nxFKu2f9goBX2O5FJ7P++MO0IquYM6DquNZVxXHcrK56uN6gKtBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1c64e4afb35f6756c924cbf368799ba3fbc7a469280c97b76332794873b64b87","last_reissued_at":"2026-07-05T04:25:26.384088Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:25:26.384088Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Reinforcement Learning from Demonstration Using Local Ensemble and Reparameterization with Split and Merge of Expert Policies","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Fang Liu, Yu Wang","submitted_at":"2022-05-23T03:36:24Z","abstract_excerpt":"The current work on reinforcement learning (RL) from demonstrations often assumes the demonstrations are samples from an optimal policy, an unrealistic assumption in practice. When demonstrations are generated by sub-optimal policies or have sparse state-action pairs, policy learned from sub-optimal demonstrations may mislead an agent with incorrect or non-local action decisions. We propose a new method called Local Ensemble and Reparameterization with Split and Merge of expert policies (LEARN-SAM) to improve efficiency and make better use of the sub-optimal demonstrations. First, LEARN-SAM em"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.11019","kind":"arxiv","version":1},"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/2205.11019/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":"2205.11019","created_at":"2026-07-05T04:25:26.384153+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.11019v1","created_at":"2026-07-05T04:25:26.384153+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.11019","created_at":"2026-07-05T04:25:26.384153+00:00"},{"alias_kind":"pith_short_12","alias_value":"DRSOJL5TL5TV","created_at":"2026-07-05T04:25:26.384153+00:00"},{"alias_kind":"pith_short_16","alias_value":"DRSOJL5TL5TVNSJE","created_at":"2026-07-05T04:25:26.384153+00:00"},{"alias_kind":"pith_short_8","alias_value":"DRSOJL5T","created_at":"2026-07-05T04:25:26.384153+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DRSOJL5TL5TVNSJEZPZWQ6M3UP","json":"https://pith.science/pith/DRSOJL5TL5TVNSJEZPZWQ6M3UP.json","graph_json":"https://pith.science/api/pith-number/DRSOJL5TL5TVNSJEZPZWQ6M3UP/graph.json","events_json":"https://pith.science/api/pith-number/DRSOJL5TL5TVNSJEZPZWQ6M3UP/events.json","paper":"https://pith.science/paper/DRSOJL5T"},"agent_actions":{"view_html":"https://pith.science/pith/DRSOJL5TL5TVNSJEZPZWQ6M3UP","download_json":"https://pith.science/pith/DRSOJL5TL5TVNSJEZPZWQ6M3UP.json","view_paper":"https://pith.science/paper/DRSOJL5T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.11019&json=true","fetch_graph":"https://pith.science/api/pith-number/DRSOJL5TL5TVNSJEZPZWQ6M3UP/graph.json","fetch_events":"https://pith.science/api/pith-number/DRSOJL5TL5TVNSJEZPZWQ6M3UP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DRSOJL5TL5TVNSJEZPZWQ6M3UP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DRSOJL5TL5TVNSJEZPZWQ6M3UP/action/storage_attestation","attest_author":"https://pith.science/pith/DRSOJL5TL5TVNSJEZPZWQ6M3UP/action/author_attestation","sign_citation":"https://pith.science/pith/DRSOJL5TL5TVNSJEZPZWQ6M3UP/action/citation_signature","submit_replication":"https://pith.science/pith/DRSOJL5TL5TVNSJEZPZWQ6M3UP/action/replication_record"}},"created_at":"2026-07-05T04:25:26.384153+00:00","updated_at":"2026-07-05T04:25:26.384153+00:00"}