{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HXEKJ36EG5347KXV2GJ53YVCKI","short_pith_number":"pith:HXEKJ36E","schema_version":"1.0","canonical_sha256":"3dc8a4efc43777cfaaf5d193dde2a2520ad2f2c9d46034febe420df209466e10","source":{"kind":"arxiv","id":"2401.05899","version":1},"attestation_state":"computed","paper":{"title":"Optimistic Model Rollouts for Pessimistic Offline Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dawei Feng, Ding Bo, Huaimin Wang, Kele Xu, Xudong Gong, Yiying Li, Yuanzhao Zhai, Zijian Gao","submitted_at":"2024-01-11T13:19:45Z","abstract_excerpt":"Model-based offline reinforcement learning (RL) has made remarkable progress, offering a promising avenue for improving generalization with synthetic model rollouts. Existing works primarily focus on incorporating pessimism for policy optimization, usually via constructing a Pessimistic Markov Decision Process (P-MDP). However, the P-MDP discourages the policies from learning in out-of-distribution (OOD) regions beyond the support of offline datasets, which can under-utilize the generalization ability of dynamics models. In contrast, we propose constructing an Optimistic MDP (O-MDP). We initia"},"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":"2401.05899","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-11T13:19:45Z","cross_cats_sorted":[],"title_canon_sha256":"506179f15c17300648844cd3f9c93b885139a2bacde4b80d226ffad9a625aa5a","abstract_canon_sha256":"4afd89eb194ab0b0794c39e004008f2c5f026e68065c19ef87380af89dec43fa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:32:37.436328Z","signature_b64":"TmSb/Rcs10rKRaJM38r+sVcpQCBG9fV6C59CGFQ7woXMYDP4zligyFHc0MRz7ZEdUMDKdE1/88ZyaM+DYzsFDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3dc8a4efc43777cfaaf5d193dde2a2520ad2f2c9d46034febe420df209466e10","last_reissued_at":"2026-07-05T07:32:37.435881Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:32:37.435881Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimistic Model Rollouts for Pessimistic Offline Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dawei Feng, Ding Bo, Huaimin Wang, Kele Xu, Xudong Gong, Yiying Li, Yuanzhao Zhai, Zijian Gao","submitted_at":"2024-01-11T13:19:45Z","abstract_excerpt":"Model-based offline reinforcement learning (RL) has made remarkable progress, offering a promising avenue for improving generalization with synthetic model rollouts. Existing works primarily focus on incorporating pessimism for policy optimization, usually via constructing a Pessimistic Markov Decision Process (P-MDP). However, the P-MDP discourages the policies from learning in out-of-distribution (OOD) regions beyond the support of offline datasets, which can under-utilize the generalization ability of dynamics models. In contrast, we propose constructing an Optimistic MDP (O-MDP). We initia"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.05899","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/2401.05899/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":"2401.05899","created_at":"2026-07-05T07:32:37.435940+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.05899v1","created_at":"2026-07-05T07:32:37.435940+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.05899","created_at":"2026-07-05T07:32:37.435940+00:00"},{"alias_kind":"pith_short_12","alias_value":"HXEKJ36EG534","created_at":"2026-07-05T07:32:37.435940+00:00"},{"alias_kind":"pith_short_16","alias_value":"HXEKJ36EG5347KXV","created_at":"2026-07-05T07:32:37.435940+00:00"},{"alias_kind":"pith_short_8","alias_value":"HXEKJ36E","created_at":"2026-07-05T07:32:37.435940+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/HXEKJ36EG5347KXV2GJ53YVCKI","json":"https://pith.science/pith/HXEKJ36EG5347KXV2GJ53YVCKI.json","graph_json":"https://pith.science/api/pith-number/HXEKJ36EG5347KXV2GJ53YVCKI/graph.json","events_json":"https://pith.science/api/pith-number/HXEKJ36EG5347KXV2GJ53YVCKI/events.json","paper":"https://pith.science/paper/HXEKJ36E"},"agent_actions":{"view_html":"https://pith.science/pith/HXEKJ36EG5347KXV2GJ53YVCKI","download_json":"https://pith.science/pith/HXEKJ36EG5347KXV2GJ53YVCKI.json","view_paper":"https://pith.science/paper/HXEKJ36E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.05899&json=true","fetch_graph":"https://pith.science/api/pith-number/HXEKJ36EG5347KXV2GJ53YVCKI/graph.json","fetch_events":"https://pith.science/api/pith-number/HXEKJ36EG5347KXV2GJ53YVCKI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HXEKJ36EG5347KXV2GJ53YVCKI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HXEKJ36EG5347KXV2GJ53YVCKI/action/storage_attestation","attest_author":"https://pith.science/pith/HXEKJ36EG5347KXV2GJ53YVCKI/action/author_attestation","sign_citation":"https://pith.science/pith/HXEKJ36EG5347KXV2GJ53YVCKI/action/citation_signature","submit_replication":"https://pith.science/pith/HXEKJ36EG5347KXV2GJ53YVCKI/action/replication_record"}},"created_at":"2026-07-05T07:32:37.435940+00:00","updated_at":"2026-07-05T07:32:37.435940+00:00"}