{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OL3D7BFXXDKIEBQSJZ7QIPLB5P","short_pith_number":"pith:OL3D7BFX","schema_version":"1.0","canonical_sha256":"72f63f84b7b8d48206124e7f043d61ebc0ea87b471cceecfe5e679f2e1c0596e","source":{"kind":"arxiv","id":"2405.19080","version":1},"attestation_state":"computed","paper":{"title":"OMPO: A Unified Framework for RL under Policy and Dynamics Shifts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Fuchun Sun, Huazhe Xu, Jianwei Zhang, Tianying Ji, Xianyuan Zhan, Yu Luo","submitted_at":"2024-05-29T13:36:36Z","abstract_excerpt":"Training reinforcement learning policies using environment interaction data collected from varying policies or dynamics presents a fundamental challenge. Existing works often overlook the distribution discrepancies induced by policy or dynamics shifts, or rely on specialized algorithms with task priors, thus often resulting in suboptimal policy performances and high learning variances. In this paper, we identify a unified strategy for online RL policy learning under diverse settings of policy and dynamics shifts: transition occupancy matching. In light of this, we introduce a surrogate policy "},"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":"2405.19080","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-29T13:36:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"79787794210844e1f10946559c1dcfec0371503a570ffa58608ea3f3f4d31336","abstract_canon_sha256":"4bbb8b9c429bdc2cb162612c6cf33c937cfe852624fb43752647c0ff5c454906"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:24:47.543723Z","signature_b64":"/9BJbSBAK9fwC/11MfzWYKA7gAN99TLHsSZ9Ep3GFbGZLJGhe92JRCk1S5RFB6O9dxpq5Baf4d33nbEuDUHfDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"72f63f84b7b8d48206124e7f043d61ebc0ea87b471cceecfe5e679f2e1c0596e","last_reissued_at":"2026-07-05T08:24:47.543225Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:24:47.543225Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OMPO: A Unified Framework for RL under Policy and Dynamics Shifts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Fuchun Sun, Huazhe Xu, Jianwei Zhang, Tianying Ji, Xianyuan Zhan, Yu Luo","submitted_at":"2024-05-29T13:36:36Z","abstract_excerpt":"Training reinforcement learning policies using environment interaction data collected from varying policies or dynamics presents a fundamental challenge. Existing works often overlook the distribution discrepancies induced by policy or dynamics shifts, or rely on specialized algorithms with task priors, thus often resulting in suboptimal policy performances and high learning variances. In this paper, we identify a unified strategy for online RL policy learning under diverse settings of policy and dynamics shifts: transition occupancy matching. In light of this, we introduce a surrogate policy "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.19080","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/2405.19080/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":"2405.19080","created_at":"2026-07-05T08:24:47.543284+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.19080v1","created_at":"2026-07-05T08:24:47.543284+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.19080","created_at":"2026-07-05T08:24:47.543284+00:00"},{"alias_kind":"pith_short_12","alias_value":"OL3D7BFXXDKI","created_at":"2026-07-05T08:24:47.543284+00:00"},{"alias_kind":"pith_short_16","alias_value":"OL3D7BFXXDKIEBQS","created_at":"2026-07-05T08:24:47.543284+00:00"},{"alias_kind":"pith_short_8","alias_value":"OL3D7BFX","created_at":"2026-07-05T08:24:47.543284+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":95,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OL3D7BFXXDKIEBQSJZ7QIPLB5P","json":"https://pith.science/pith/OL3D7BFXXDKIEBQSJZ7QIPLB5P.json","graph_json":"https://pith.science/api/pith-number/OL3D7BFXXDKIEBQSJZ7QIPLB5P/graph.json","events_json":"https://pith.science/api/pith-number/OL3D7BFXXDKIEBQSJZ7QIPLB5P/events.json","paper":"https://pith.science/paper/OL3D7BFX"},"agent_actions":{"view_html":"https://pith.science/pith/OL3D7BFXXDKIEBQSJZ7QIPLB5P","download_json":"https://pith.science/pith/OL3D7BFXXDKIEBQSJZ7QIPLB5P.json","view_paper":"https://pith.science/paper/OL3D7BFX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.19080&json=true","fetch_graph":"https://pith.science/api/pith-number/OL3D7BFXXDKIEBQSJZ7QIPLB5P/graph.json","fetch_events":"https://pith.science/api/pith-number/OL3D7BFXXDKIEBQSJZ7QIPLB5P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OL3D7BFXXDKIEBQSJZ7QIPLB5P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OL3D7BFXXDKIEBQSJZ7QIPLB5P/action/storage_attestation","attest_author":"https://pith.science/pith/OL3D7BFXXDKIEBQSJZ7QIPLB5P/action/author_attestation","sign_citation":"https://pith.science/pith/OL3D7BFXXDKIEBQSJZ7QIPLB5P/action/citation_signature","submit_replication":"https://pith.science/pith/OL3D7BFXXDKIEBQSJZ7QIPLB5P/action/replication_record"}},"created_at":"2026-07-05T08:24:47.543284+00:00","updated_at":"2026-07-05T08:24:47.543284+00:00"}