{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:FOQPOZQ3M62RP6W6727JJST5PC","short_pith_number":"pith:FOQPOZQ3","schema_version":"1.0","canonical_sha256":"2ba0f7661b67b517fadefebe94ca7d788d48a85357da98e3ce5d802499599418","source":{"kind":"arxiv","id":"2007.12306","version":4},"attestation_state":"computed","paper":{"title":"Value-Decomposition Multi-Agent Actor-Critics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.MA"],"primary_cat":"cs.AI","authors_text":"Jianyu Su, Peter A. Beling, Stephen Adams","submitted_at":"2020-07-24T00:50:02Z","abstract_excerpt":"The exploitation of extra state information has been an active research area in multi-agent reinforcement learning (MARL). QMIX represents the joint action-value using a non-negative function approximator and achieves the best performance, by far, on multi-agent benchmarks, StarCraft II micromanagement tasks. However, our experiments show that, in some cases, QMIX is incompatible with A2C, a training paradigm that promotes algorithm training efficiency. To obtain a reasonable trade-off between training efficiency and algorithm performance, we extend value-decomposition to actor-critics that ar"},"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":"2007.12306","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2020-07-24T00:50:02Z","cross_cats_sorted":["cs.LG","cs.MA"],"title_canon_sha256":"8f6c854fa52e8aafe702a5ca3ed9ea80a235fdca12472c58a82a44a31e3c23d5","abstract_canon_sha256":"c43129f36b3bdfadb49452140422fadc3d8c1cd35890832dada36839987c8f41"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:00:31.309262Z","signature_b64":"47ABhhMuhDMoI12ox492dDIfviUFPLetwsojOo1zxega5k3WxRXPl42OOv0aySACaqJchrbDNMxBBAUZb6yNCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2ba0f7661b67b517fadefebe94ca7d788d48a85357da98e3ce5d802499599418","last_reissued_at":"2026-07-05T02:00:31.308770Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:00:31.308770Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Value-Decomposition Multi-Agent Actor-Critics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.MA"],"primary_cat":"cs.AI","authors_text":"Jianyu Su, Peter A. Beling, Stephen Adams","submitted_at":"2020-07-24T00:50:02Z","abstract_excerpt":"The exploitation of extra state information has been an active research area in multi-agent reinforcement learning (MARL). QMIX represents the joint action-value using a non-negative function approximator and achieves the best performance, by far, on multi-agent benchmarks, StarCraft II micromanagement tasks. However, our experiments show that, in some cases, QMIX is incompatible with A2C, a training paradigm that promotes algorithm training efficiency. To obtain a reasonable trade-off between training efficiency and algorithm performance, we extend value-decomposition to actor-critics that ar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.12306","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/2007.12306/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":"2007.12306","created_at":"2026-07-05T02:00:31.308835+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.12306v4","created_at":"2026-07-05T02:00:31.308835+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.12306","created_at":"2026-07-05T02:00:31.308835+00:00"},{"alias_kind":"pith_short_12","alias_value":"FOQPOZQ3M62R","created_at":"2026-07-05T02:00:31.308835+00:00"},{"alias_kind":"pith_short_16","alias_value":"FOQPOZQ3M62RP6W6","created_at":"2026-07-05T02:00:31.308835+00:00"},{"alias_kind":"pith_short_8","alias_value":"FOQPOZQ3","created_at":"2026-07-05T02:00:31.308835+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.03723","citing_title":"Speaking the Language of Teamwork: LLM-Guided Credit Assignment in Multi-Agent Reinforcement Learning","ref_index":9,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FOQPOZQ3M62RP6W6727JJST5PC","json":"https://pith.science/pith/FOQPOZQ3M62RP6W6727JJST5PC.json","graph_json":"https://pith.science/api/pith-number/FOQPOZQ3M62RP6W6727JJST5PC/graph.json","events_json":"https://pith.science/api/pith-number/FOQPOZQ3M62RP6W6727JJST5PC/events.json","paper":"https://pith.science/paper/FOQPOZQ3"},"agent_actions":{"view_html":"https://pith.science/pith/FOQPOZQ3M62RP6W6727JJST5PC","download_json":"https://pith.science/pith/FOQPOZQ3M62RP6W6727JJST5PC.json","view_paper":"https://pith.science/paper/FOQPOZQ3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.12306&json=true","fetch_graph":"https://pith.science/api/pith-number/FOQPOZQ3M62RP6W6727JJST5PC/graph.json","fetch_events":"https://pith.science/api/pith-number/FOQPOZQ3M62RP6W6727JJST5PC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FOQPOZQ3M62RP6W6727JJST5PC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FOQPOZQ3M62RP6W6727JJST5PC/action/storage_attestation","attest_author":"https://pith.science/pith/FOQPOZQ3M62RP6W6727JJST5PC/action/author_attestation","sign_citation":"https://pith.science/pith/FOQPOZQ3M62RP6W6727JJST5PC/action/citation_signature","submit_replication":"https://pith.science/pith/FOQPOZQ3M62RP6W6727JJST5PC/action/replication_record"}},"created_at":"2026-07-05T02:00:31.308835+00:00","updated_at":"2026-07-05T02:00:31.308835+00:00"}