{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RHCLHS4I3ZDA7TMDT6FADOBI63","short_pith_number":"pith:RHCLHS4I","schema_version":"1.0","canonical_sha256":"89c4b3cb88de460fcd839f8a01b828f6c6d42ff80d5765be330de4c54c623572","source":{"kind":"arxiv","id":"2401.08819","version":2},"attestation_state":"computed","paper":{"title":"Learning from Sparse Offline Datasets via Conservative Density Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ding Zhao, Henry Lam, Yihang Yao, Zhepeng Cen, Zitong Wang, Zuxin Liu","submitted_at":"2024-01-16T20:42:15Z","abstract_excerpt":"Offline reinforcement learning (RL) offers a promising direction for learning policies from pre-collected datasets without requiring further interactions with the environment. However, existing methods struggle to handle out-of-distribution (OOD) extrapolation errors, especially in sparse reward or scarce data settings. In this paper, we propose a novel training algorithm called Conservative Density Estimation (CDE), which addresses this challenge by explicitly imposing constraints on the state-action occupancy stationary distribution. CDE overcomes the limitations of existing approaches, such"},"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":"2401.08819","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-16T20:42:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ef59f0bf018c9cc66ba018c7a29350bc6e96c2c18528c98ba15f7bc5e8422c00","abstract_canon_sha256":"1da091f6f3fe4e7f779425cce59b029e22a3ad372e83f6d80d38c3b642f3b6de"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:54:35.424647Z","signature_b64":"0xLEnEbnHgymph3GvJZy3kfzmerg5+GnhZL3iOAbuIJS0oc3UVhjFBCpDZgjAIBia1a3p9/OXsBfyiTr72vcCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"89c4b3cb88de460fcd839f8a01b828f6c6d42ff80d5765be330de4c54c623572","last_reissued_at":"2026-07-05T07:54:35.424151Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:54:35.424151Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning from Sparse Offline Datasets via Conservative Density Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ding Zhao, Henry Lam, Yihang Yao, Zhepeng Cen, Zitong Wang, Zuxin Liu","submitted_at":"2024-01-16T20:42:15Z","abstract_excerpt":"Offline reinforcement learning (RL) offers a promising direction for learning policies from pre-collected datasets without requiring further interactions with the environment. However, existing methods struggle to handle out-of-distribution (OOD) extrapolation errors, especially in sparse reward or scarce data settings. In this paper, we propose a novel training algorithm called Conservative Density Estimation (CDE), which addresses this challenge by explicitly imposing constraints on the state-action occupancy stationary distribution. CDE overcomes the limitations of existing approaches, such"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.08819","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/2401.08819/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":"2401.08819","created_at":"2026-07-05T07:54:35.424209+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.08819v2","created_at":"2026-07-05T07:54:35.424209+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.08819","created_at":"2026-07-05T07:54:35.424209+00:00"},{"alias_kind":"pith_short_12","alias_value":"RHCLHS4I3ZDA","created_at":"2026-07-05T07:54:35.424209+00:00"},{"alias_kind":"pith_short_16","alias_value":"RHCLHS4I3ZDA7TMD","created_at":"2026-07-05T07:54:35.424209+00:00"},{"alias_kind":"pith_short_8","alias_value":"RHCLHS4I","created_at":"2026-07-05T07:54:35.424209+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/RHCLHS4I3ZDA7TMDT6FADOBI63","json":"https://pith.science/pith/RHCLHS4I3ZDA7TMDT6FADOBI63.json","graph_json":"https://pith.science/api/pith-number/RHCLHS4I3ZDA7TMDT6FADOBI63/graph.json","events_json":"https://pith.science/api/pith-number/RHCLHS4I3ZDA7TMDT6FADOBI63/events.json","paper":"https://pith.science/paper/RHCLHS4I"},"agent_actions":{"view_html":"https://pith.science/pith/RHCLHS4I3ZDA7TMDT6FADOBI63","download_json":"https://pith.science/pith/RHCLHS4I3ZDA7TMDT6FADOBI63.json","view_paper":"https://pith.science/paper/RHCLHS4I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.08819&json=true","fetch_graph":"https://pith.science/api/pith-number/RHCLHS4I3ZDA7TMDT6FADOBI63/graph.json","fetch_events":"https://pith.science/api/pith-number/RHCLHS4I3ZDA7TMDT6FADOBI63/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RHCLHS4I3ZDA7TMDT6FADOBI63/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RHCLHS4I3ZDA7TMDT6FADOBI63/action/storage_attestation","attest_author":"https://pith.science/pith/RHCLHS4I3ZDA7TMDT6FADOBI63/action/author_attestation","sign_citation":"https://pith.science/pith/RHCLHS4I3ZDA7TMDT6FADOBI63/action/citation_signature","submit_replication":"https://pith.science/pith/RHCLHS4I3ZDA7TMDT6FADOBI63/action/replication_record"}},"created_at":"2026-07-05T07:54:35.424209+00:00","updated_at":"2026-07-05T07:54:35.424209+00:00"}