{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:U73NBYEYH6XZCQH5CFHIZCZTWR","short_pith_number":"pith:U73NBYEY","schema_version":"1.0","canonical_sha256":"a7f6d0e0983faf9140fd114e8c8b33b476eab31b3851c79a048151ab8014bb7a","source":{"kind":"arxiv","id":"2210.08349","version":4},"attestation_state":"computed","paper":{"title":"When to Update Your Model: Constrained Model-based Reinforcement Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Fengxiang He, Fuchun Sun, Mingxuan Jing, Tianying Ji, Wenbing Huang, Yu Luo","submitted_at":"2022-10-15T17:57:43Z","abstract_excerpt":"Designing and analyzing model-based RL (MBRL) algorithms with guaranteed monotonic improvement has been challenging, mainly due to the interdependence between policy optimization and model learning. Existing discrepancy bounds generally ignore the impacts of model shifts, and their corresponding algorithms are prone to degrade performance by drastic model updating. In this work, we first propose a novel and general theoretical scheme for a non-decreasing performance guarantee of MBRL. Our follow-up derived bounds reveal the relationship between model shifts and performance improvement. These d"},"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":"2210.08349","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2022-10-15T17:57:43Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ebb96d36e1ddf1b5e2d0270590fdf25d517a1b8c7a7ff02df2a434d5ce4579f0","abstract_canon_sha256":"9b102f0697cc86f001c28d185dcc657c58d556b7178c0beeee40e0e2a921f0a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:10:21.234645Z","signature_b64":"ParQMseeysDMTnozegHRXqba0qcJgrqPv/zKYr3nvl+4gGtj+gPAvKZWFDceg7SozhQYcXMYQxFi96GghJfgBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a7f6d0e0983faf9140fd114e8c8b33b476eab31b3851c79a048151ab8014bb7a","last_reissued_at":"2026-07-05T07:10:21.234132Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:10:21.234132Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When to Update Your Model: Constrained Model-based Reinforcement Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Fengxiang He, Fuchun Sun, Mingxuan Jing, Tianying Ji, Wenbing Huang, Yu Luo","submitted_at":"2022-10-15T17:57:43Z","abstract_excerpt":"Designing and analyzing model-based RL (MBRL) algorithms with guaranteed monotonic improvement has been challenging, mainly due to the interdependence between policy optimization and model learning. Existing discrepancy bounds generally ignore the impacts of model shifts, and their corresponding algorithms are prone to degrade performance by drastic model updating. In this work, we first propose a novel and general theoretical scheme for a non-decreasing performance guarantee of MBRL. Our follow-up derived bounds reveal the relationship between model shifts and performance improvement. These d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.08349","kind":"arxiv","version":4},"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/2210.08349/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":"2210.08349","created_at":"2026-07-05T07:10:21.234190+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.08349v4","created_at":"2026-07-05T07:10:21.234190+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.08349","created_at":"2026-07-05T07:10:21.234190+00:00"},{"alias_kind":"pith_short_12","alias_value":"U73NBYEYH6XZ","created_at":"2026-07-05T07:10:21.234190+00:00"},{"alias_kind":"pith_short_16","alias_value":"U73NBYEYH6XZCQH5","created_at":"2026-07-05T07:10:21.234190+00:00"},{"alias_kind":"pith_short_8","alias_value":"U73NBYEY","created_at":"2026-07-05T07:10:21.234190+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.06497","citing_title":"Hyperfastrl: Hypernetwork-based reinforcement learning for unified control of parametric chaotic PDEs","ref_index":90,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/U73NBYEYH6XZCQH5CFHIZCZTWR","json":"https://pith.science/pith/U73NBYEYH6XZCQH5CFHIZCZTWR.json","graph_json":"https://pith.science/api/pith-number/U73NBYEYH6XZCQH5CFHIZCZTWR/graph.json","events_json":"https://pith.science/api/pith-number/U73NBYEYH6XZCQH5CFHIZCZTWR/events.json","paper":"https://pith.science/paper/U73NBYEY"},"agent_actions":{"view_html":"https://pith.science/pith/U73NBYEYH6XZCQH5CFHIZCZTWR","download_json":"https://pith.science/pith/U73NBYEYH6XZCQH5CFHIZCZTWR.json","view_paper":"https://pith.science/paper/U73NBYEY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.08349&json=true","fetch_graph":"https://pith.science/api/pith-number/U73NBYEYH6XZCQH5CFHIZCZTWR/graph.json","fetch_events":"https://pith.science/api/pith-number/U73NBYEYH6XZCQH5CFHIZCZTWR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U73NBYEYH6XZCQH5CFHIZCZTWR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U73NBYEYH6XZCQH5CFHIZCZTWR/action/storage_attestation","attest_author":"https://pith.science/pith/U73NBYEYH6XZCQH5CFHIZCZTWR/action/author_attestation","sign_citation":"https://pith.science/pith/U73NBYEYH6XZCQH5CFHIZCZTWR/action/citation_signature","submit_replication":"https://pith.science/pith/U73NBYEYH6XZCQH5CFHIZCZTWR/action/replication_record"}},"created_at":"2026-07-05T07:10:21.234190+00:00","updated_at":"2026-07-05T07:10:21.234190+00:00"}