{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QWX3NSVHDUDTHYUYOXJEESKL6G","short_pith_number":"pith:QWX3NSVH","schema_version":"1.0","canonical_sha256":"85afb6caa71d0733e29875d242494bf1b6460c532bb5d20ded8b7a369a366d18","source":{"kind":"arxiv","id":"2301.11426","version":1},"attestation_state":"computed","paper":{"title":"Model-based Offline Reinforcement Learning with Local Misspecification","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Allen Nie, Emma Brunskill, Kefan Dong, Yannis Flet-Berliac","submitted_at":"2023-01-26T21:26:56Z","abstract_excerpt":"We present a model-based offline reinforcement learning policy performance lower bound that explicitly captures dynamics model misspecification and distribution mismatch and we propose an empirical algorithm for optimal offline policy selection. Theoretically, we prove a novel safe policy improvement theorem by establishing pessimism approximations to the value function. Our key insight is to jointly consider selecting over dynamics models and policies: as long as a dynamics model can accurately represent the dynamics of the state-action pairs visited by a given policy, it is possible to appro"},"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":"2301.11426","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2023-01-26T21:26:56Z","cross_cats_sorted":[],"title_canon_sha256":"063fad1ad61443eb5b136a3ff96910c3069d18d44cdca82781679bbecca68c89","abstract_canon_sha256":"ac30e25f5701ee2841fb66812828c887d4bb054ce34d308f90a7e5e29b5b2093"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:36:21.886459Z","signature_b64":"/UjhYtWPjfNiCFtH4z0Fu6p7KKnvQJCIrdypQ7hmDoOpzNe13FCGe1eXnBNjF8Tc0ftJ1Zl73491KHAKKcdpBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"85afb6caa71d0733e29875d242494bf1b6460c532bb5d20ded8b7a369a366d18","last_reissued_at":"2026-07-05T05:36:21.885995Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:36:21.885995Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model-based Offline Reinforcement Learning with Local Misspecification","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Allen Nie, Emma Brunskill, Kefan Dong, Yannis Flet-Berliac","submitted_at":"2023-01-26T21:26:56Z","abstract_excerpt":"We present a model-based offline reinforcement learning policy performance lower bound that explicitly captures dynamics model misspecification and distribution mismatch and we propose an empirical algorithm for optimal offline policy selection. Theoretically, we prove a novel safe policy improvement theorem by establishing pessimism approximations to the value function. Our key insight is to jointly consider selecting over dynamics models and policies: as long as a dynamics model can accurately represent the dynamics of the state-action pairs visited by a given policy, it is possible to appro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.11426","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/2301.11426/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":"2301.11426","created_at":"2026-07-05T05:36:21.886046+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.11426v1","created_at":"2026-07-05T05:36:21.886046+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.11426","created_at":"2026-07-05T05:36:21.886046+00:00"},{"alias_kind":"pith_short_12","alias_value":"QWX3NSVHDUDT","created_at":"2026-07-05T05:36:21.886046+00:00"},{"alias_kind":"pith_short_16","alias_value":"QWX3NSVHDUDTHYUY","created_at":"2026-07-05T05:36:21.886046+00:00"},{"alias_kind":"pith_short_8","alias_value":"QWX3NSVH","created_at":"2026-07-05T05:36:21.886046+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/QWX3NSVHDUDTHYUYOXJEESKL6G","json":"https://pith.science/pith/QWX3NSVHDUDTHYUYOXJEESKL6G.json","graph_json":"https://pith.science/api/pith-number/QWX3NSVHDUDTHYUYOXJEESKL6G/graph.json","events_json":"https://pith.science/api/pith-number/QWX3NSVHDUDTHYUYOXJEESKL6G/events.json","paper":"https://pith.science/paper/QWX3NSVH"},"agent_actions":{"view_html":"https://pith.science/pith/QWX3NSVHDUDTHYUYOXJEESKL6G","download_json":"https://pith.science/pith/QWX3NSVHDUDTHYUYOXJEESKL6G.json","view_paper":"https://pith.science/paper/QWX3NSVH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.11426&json=true","fetch_graph":"https://pith.science/api/pith-number/QWX3NSVHDUDTHYUYOXJEESKL6G/graph.json","fetch_events":"https://pith.science/api/pith-number/QWX3NSVHDUDTHYUYOXJEESKL6G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QWX3NSVHDUDTHYUYOXJEESKL6G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QWX3NSVHDUDTHYUYOXJEESKL6G/action/storage_attestation","attest_author":"https://pith.science/pith/QWX3NSVHDUDTHYUYOXJEESKL6G/action/author_attestation","sign_citation":"https://pith.science/pith/QWX3NSVHDUDTHYUYOXJEESKL6G/action/citation_signature","submit_replication":"https://pith.science/pith/QWX3NSVHDUDTHYUYOXJEESKL6G/action/replication_record"}},"created_at":"2026-07-05T05:36:21.886046+00:00","updated_at":"2026-07-05T05:36:21.886046+00:00"}