{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:XSEVWGQ53QKDIK2XZZIY74DIKO","short_pith_number":"pith:XSEVWGQ5","schema_version":"1.0","canonical_sha256":"bc895b1a1ddc14342b57ce518ff06853867eb709c313721acb251031a340064c","source":{"kind":"arxiv","id":"2411.00666","version":1},"attestation_state":"computed","paper":{"title":"Beyond the Boundaries of Proximal Policy Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Benjamin Ellis, Charlie B. Tan, Edan Toledo, Ferenc Husz\\'ar, Jakob N. Foerster","submitted_at":"2024-11-01T15:29:10Z","abstract_excerpt":"Proximal policy optimization (PPO) is a widely-used algorithm for on-policy reinforcement learning. This work offers an alternative perspective of PPO, in which it is decomposed into the inner-loop estimation of update vectors, and the outer-loop application of updates using gradient ascent with unity learning rate. Using this insight we propose outer proximal policy optimization (outer-PPO); a framework wherein these update vectors are applied using an arbitrary gradient-based optimizer. The decoupling of update estimation and update application enabled by outer-PPO highlights several implici"},"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":"2411.00666","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-01T15:29:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"93e01bf3c60f08f707c87f17cdfbd1bd016bf4b28952430808dc540a4b8ba385","abstract_canon_sha256":"ce4ec11f79205a9ae1c1fca5280203862ee87d608e9ee1d091c163a3cac0051f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:29:46.491045Z","signature_b64":"A0FEawnPaUHRUxVAdSoh54fFsaTIXApPuI7DR3Egw0/YwojekfM1BdhAoAWm5+5hrArqVsIB+R8D92mZ5al6BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc895b1a1ddc14342b57ce518ff06853867eb709c313721acb251031a340064c","last_reissued_at":"2026-07-05T09:29:46.490555Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:29:46.490555Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond the Boundaries of Proximal Policy Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Benjamin Ellis, Charlie B. Tan, Edan Toledo, Ferenc Husz\\'ar, Jakob N. Foerster","submitted_at":"2024-11-01T15:29:10Z","abstract_excerpt":"Proximal policy optimization (PPO) is a widely-used algorithm for on-policy reinforcement learning. This work offers an alternative perspective of PPO, in which it is decomposed into the inner-loop estimation of update vectors, and the outer-loop application of updates using gradient ascent with unity learning rate. Using this insight we propose outer proximal policy optimization (outer-PPO); a framework wherein these update vectors are applied using an arbitrary gradient-based optimizer. The decoupling of update estimation and update application enabled by outer-PPO highlights several implici"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.00666","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/2411.00666/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":"2411.00666","created_at":"2026-07-05T09:29:46.490611+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.00666v1","created_at":"2026-07-05T09:29:46.490611+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.00666","created_at":"2026-07-05T09:29:46.490611+00:00"},{"alias_kind":"pith_short_12","alias_value":"XSEVWGQ53QKD","created_at":"2026-07-05T09:29:46.490611+00:00"},{"alias_kind":"pith_short_16","alias_value":"XSEVWGQ53QKDIK2X","created_at":"2026-07-05T09:29:46.490611+00:00"},{"alias_kind":"pith_short_8","alias_value":"XSEVWGQ5","created_at":"2026-07-05T09:29:46.490611+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.18578","citing_title":"Bounded Ratio Reinforcement Learning","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XSEVWGQ53QKDIK2XZZIY74DIKO","json":"https://pith.science/pith/XSEVWGQ53QKDIK2XZZIY74DIKO.json","graph_json":"https://pith.science/api/pith-number/XSEVWGQ53QKDIK2XZZIY74DIKO/graph.json","events_json":"https://pith.science/api/pith-number/XSEVWGQ53QKDIK2XZZIY74DIKO/events.json","paper":"https://pith.science/paper/XSEVWGQ5"},"agent_actions":{"view_html":"https://pith.science/pith/XSEVWGQ53QKDIK2XZZIY74DIKO","download_json":"https://pith.science/pith/XSEVWGQ53QKDIK2XZZIY74DIKO.json","view_paper":"https://pith.science/paper/XSEVWGQ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.00666&json=true","fetch_graph":"https://pith.science/api/pith-number/XSEVWGQ53QKDIK2XZZIY74DIKO/graph.json","fetch_events":"https://pith.science/api/pith-number/XSEVWGQ53QKDIK2XZZIY74DIKO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XSEVWGQ53QKDIK2XZZIY74DIKO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XSEVWGQ53QKDIK2XZZIY74DIKO/action/storage_attestation","attest_author":"https://pith.science/pith/XSEVWGQ53QKDIK2XZZIY74DIKO/action/author_attestation","sign_citation":"https://pith.science/pith/XSEVWGQ53QKDIK2XZZIY74DIKO/action/citation_signature","submit_replication":"https://pith.science/pith/XSEVWGQ53QKDIK2XZZIY74DIKO/action/replication_record"}},"created_at":"2026-07-05T09:29:46.490611+00:00","updated_at":"2026-07-05T09:29:46.490611+00:00"}