{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MPWU3WJ4LUMILHFDALZZDEA4ZO","short_pith_number":"pith:MPWU3WJ4","schema_version":"1.0","canonical_sha256":"63ed4dd93c5d18859ca302f391901ccb90ae71316151ced117703f4b013f43dc","source":{"kind":"arxiv","id":"2404.08003","version":5},"attestation_state":"computed","paper":{"title":"Asynchronous Federated Reinforcement Learning with Policy Gradient Updates: Algorithm Design and Convergence Analysis","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC","cs.NI"],"primary_cat":"cs.LG","authors_text":"Abolfazl Hashemi, Christopher G. Brinton, Dong-Jun Han, Guangchen Lan, Vaneet Aggarwal","submitted_at":"2024-04-09T04:21:13Z","abstract_excerpt":"To improve the efficiency of reinforcement learning (RL), we propose a novel asynchronous federated reinforcement learning (FedRL) framework termed AFedPG, which constructs a global model through collaboration among $N$ agents using policy gradient (PG) updates. To address the challenge of lagged policies in asynchronous settings, we design a delay-adaptive lookahead technique \\textit{specifically for FedRL} that can effectively handle heterogeneous arrival times of policy gradients. We analyze the theoretical global convergence bound of AFedPG, and characterize the advantage of the proposed a"},"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":"2404.08003","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-09T04:21:13Z","cross_cats_sorted":["cs.DC","cs.NI"],"title_canon_sha256":"91e9d0448fc0ee066f906864984790cae657788da20c931cf8650dbf2d2a6531","abstract_canon_sha256":"1c7ada37838543a3e5dfcb66771a257a740c09af0a13866662b9083d5ed4c0af"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:04:38.289366Z","signature_b64":"aBQtS3ywMKfNI2b5eAPJe/77xmJiG27ahATiJjf2aqrrDDFzVUp7cPY5KKK5PQSgbob2CfXlQ2orYKxS5AJnDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"63ed4dd93c5d18859ca302f391901ccb90ae71316151ced117703f4b013f43dc","last_reissued_at":"2026-07-05T10:04:38.288816Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:04:38.288816Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Asynchronous Federated Reinforcement Learning with Policy Gradient Updates: Algorithm Design and Convergence Analysis","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC","cs.NI"],"primary_cat":"cs.LG","authors_text":"Abolfazl Hashemi, Christopher G. Brinton, Dong-Jun Han, Guangchen Lan, Vaneet Aggarwal","submitted_at":"2024-04-09T04:21:13Z","abstract_excerpt":"To improve the efficiency of reinforcement learning (RL), we propose a novel asynchronous federated reinforcement learning (FedRL) framework termed AFedPG, which constructs a global model through collaboration among $N$ agents using policy gradient (PG) updates. To address the challenge of lagged policies in asynchronous settings, we design a delay-adaptive lookahead technique \\textit{specifically for FedRL} that can effectively handle heterogeneous arrival times of policy gradients. We analyze the theoretical global convergence bound of AFedPG, and characterize the advantage of the proposed a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.08003","kind":"arxiv","version":5},"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/2404.08003/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":"2404.08003","created_at":"2026-07-05T10:04:38.288877+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.08003v5","created_at":"2026-07-05T10:04:38.288877+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.08003","created_at":"2026-07-05T10:04:38.288877+00:00"},{"alias_kind":"pith_short_12","alias_value":"MPWU3WJ4LUMI","created_at":"2026-07-05T10:04:38.288877+00:00"},{"alias_kind":"pith_short_16","alias_value":"MPWU3WJ4LUMILHFD","created_at":"2026-07-05T10:04:38.288877+00:00"},{"alias_kind":"pith_short_8","alias_value":"MPWU3WJ4","created_at":"2026-07-05T10:04:38.288877+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.16690","citing_title":"UB-SMoE: Universally Balanced Sparse Mixture-of-Experts for Resource-adaptive Federated Fine-tuning of Foundation Models","ref_index":72,"is_internal_anchor":false},{"citing_arxiv_id":"2512.05372","citing_title":"Breaking the Capacity Bottleneck in Model-Heterogeneous Federated Learning via Gradual Model Restoration","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MPWU3WJ4LUMILHFDALZZDEA4ZO","json":"https://pith.science/pith/MPWU3WJ4LUMILHFDALZZDEA4ZO.json","graph_json":"https://pith.science/api/pith-number/MPWU3WJ4LUMILHFDALZZDEA4ZO/graph.json","events_json":"https://pith.science/api/pith-number/MPWU3WJ4LUMILHFDALZZDEA4ZO/events.json","paper":"https://pith.science/paper/MPWU3WJ4"},"agent_actions":{"view_html":"https://pith.science/pith/MPWU3WJ4LUMILHFDALZZDEA4ZO","download_json":"https://pith.science/pith/MPWU3WJ4LUMILHFDALZZDEA4ZO.json","view_paper":"https://pith.science/paper/MPWU3WJ4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.08003&json=true","fetch_graph":"https://pith.science/api/pith-number/MPWU3WJ4LUMILHFDALZZDEA4ZO/graph.json","fetch_events":"https://pith.science/api/pith-number/MPWU3WJ4LUMILHFDALZZDEA4ZO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MPWU3WJ4LUMILHFDALZZDEA4ZO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MPWU3WJ4LUMILHFDALZZDEA4ZO/action/storage_attestation","attest_author":"https://pith.science/pith/MPWU3WJ4LUMILHFDALZZDEA4ZO/action/author_attestation","sign_citation":"https://pith.science/pith/MPWU3WJ4LUMILHFDALZZDEA4ZO/action/citation_signature","submit_replication":"https://pith.science/pith/MPWU3WJ4LUMILHFDALZZDEA4ZO/action/replication_record"}},"created_at":"2026-07-05T10:04:38.288877+00:00","updated_at":"2026-07-05T10:04:38.288877+00:00"}