{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:MJNDURVDTQDAHYI2HRDJNGQSCQ","short_pith_number":"pith:MJNDURVD","schema_version":"1.0","canonical_sha256":"625a3a46a39c0603e11a3c46969a12143f15555fac94a7485d07f2ee88cc68fe","source":{"kind":"arxiv","id":"2007.12322","version":2},"attestation_state":"computed","paper":{"title":"Off-Policy Multi-Agent Decomposed Policy Gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.MA","stat.ML"],"primary_cat":"cs.LG","authors_text":"Beining Han, Chongjie Zhang, Heng Dong, Tonghan Wang, Yihan Wang","submitted_at":"2020-07-24T02:21:55Z","abstract_excerpt":"Multi-agent policy gradient (MAPG) methods recently witness vigorous progress. However, there is a significant performance discrepancy between MAPG methods and state-of-the-art multi-agent value-based approaches. In this paper, we investigate causes that hinder the performance of MAPG algorithms and present a multi-agent decomposed policy gradient method (DOP). This method introduces the idea of value function decomposition into the multi-agent actor-critic framework. Based on this idea, DOP supports efficient off-policy learning and addresses the issue of centralized-decentralized mismatch an"},"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":"2007.12322","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-07-24T02:21:55Z","cross_cats_sorted":["cs.MA","stat.ML"],"title_canon_sha256":"77c9657552743e11ce70c752d566c1bbc0b31db5484dcc44069253d6273ece81","abstract_canon_sha256":"4cac72b03e22ed6c419923f1621fa6b9f7c980a206cfd71a6c0d21c96d2187be"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:40:00.151194Z","signature_b64":"wQsSHO97prVmG1VMPy6PwbpniSuRmBz3aGvdMscvxbpPEIx/yFq2Ztrh6roce/uVYTG2dkTn1Vyr1ZRHnR/3AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"625a3a46a39c0603e11a3c46969a12143f15555fac94a7485d07f2ee88cc68fe","last_reissued_at":"2026-07-05T01:40:00.150810Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:40:00.150810Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Off-Policy Multi-Agent Decomposed Policy Gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.MA","stat.ML"],"primary_cat":"cs.LG","authors_text":"Beining Han, Chongjie Zhang, Heng Dong, Tonghan Wang, Yihan Wang","submitted_at":"2020-07-24T02:21:55Z","abstract_excerpt":"Multi-agent policy gradient (MAPG) methods recently witness vigorous progress. However, there is a significant performance discrepancy between MAPG methods and state-of-the-art multi-agent value-based approaches. In this paper, we investigate causes that hinder the performance of MAPG algorithms and present a multi-agent decomposed policy gradient method (DOP). This method introduces the idea of value function decomposition into the multi-agent actor-critic framework. Based on this idea, DOP supports efficient off-policy learning and addresses the issue of centralized-decentralized mismatch an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.12322","kind":"arxiv","version":2},"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/2007.12322/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":"2007.12322","created_at":"2026-07-05T01:40:00.150866+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.12322v2","created_at":"2026-07-05T01:40:00.150866+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.12322","created_at":"2026-07-05T01:40:00.150866+00:00"},{"alias_kind":"pith_short_12","alias_value":"MJNDURVDTQDA","created_at":"2026-07-05T01:40:00.150866+00:00"},{"alias_kind":"pith_short_16","alias_value":"MJNDURVDTQDAHYI2","created_at":"2026-07-05T01:40:00.150866+00:00"},{"alias_kind":"pith_short_8","alias_value":"MJNDURVD","created_at":"2026-07-05T01:40:00.150866+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.24113","citing_title":"Distributed Neural Policy Gradient Algorithm for Global Convergence of Networked Multi-Agent Reinforcement Learning","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MJNDURVDTQDAHYI2HRDJNGQSCQ","json":"https://pith.science/pith/MJNDURVDTQDAHYI2HRDJNGQSCQ.json","graph_json":"https://pith.science/api/pith-number/MJNDURVDTQDAHYI2HRDJNGQSCQ/graph.json","events_json":"https://pith.science/api/pith-number/MJNDURVDTQDAHYI2HRDJNGQSCQ/events.json","paper":"https://pith.science/paper/MJNDURVD"},"agent_actions":{"view_html":"https://pith.science/pith/MJNDURVDTQDAHYI2HRDJNGQSCQ","download_json":"https://pith.science/pith/MJNDURVDTQDAHYI2HRDJNGQSCQ.json","view_paper":"https://pith.science/paper/MJNDURVD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.12322&json=true","fetch_graph":"https://pith.science/api/pith-number/MJNDURVDTQDAHYI2HRDJNGQSCQ/graph.json","fetch_events":"https://pith.science/api/pith-number/MJNDURVDTQDAHYI2HRDJNGQSCQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MJNDURVDTQDAHYI2HRDJNGQSCQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MJNDURVDTQDAHYI2HRDJNGQSCQ/action/storage_attestation","attest_author":"https://pith.science/pith/MJNDURVDTQDAHYI2HRDJNGQSCQ/action/author_attestation","sign_citation":"https://pith.science/pith/MJNDURVDTQDAHYI2HRDJNGQSCQ/action/citation_signature","submit_replication":"https://pith.science/pith/MJNDURVDTQDAHYI2HRDJNGQSCQ/action/replication_record"}},"created_at":"2026-07-05T01:40:00.150866+00:00","updated_at":"2026-07-05T01:40:00.150866+00:00"}