{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:E3P4PPIQD5KDUJBB5YJ223F6LL","short_pith_number":"pith:E3P4PPIQ","schema_version":"1.0","canonical_sha256":"26dfc7bd101f543a2421ee13ad6cbe5afac8542477690bc1555ddf458d87eb4e","source":{"kind":"arxiv","id":"2105.03363","version":3},"attestation_state":"computed","paper":{"title":"Model-based Multi-agent Policy Optimization with Adaptive Opponent-wise Rollouts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.LG","authors_text":"Jian Shen, Ming Zhou, Weinan Zhang, Xihuai Wang","submitted_at":"2021-05-07T16:20:22Z","abstract_excerpt":"This paper investigates the model-based methods in multi-agent reinforcement learning (MARL). We specify the dynamics sample complexity and the opponent sample complexity in MARL, and conduct a theoretic analysis of return discrepancy upper bound. To reduce the upper bound with the intention of low sample complexity during the whole learning process, we propose a novel decentralized model-based MARL method, named Adaptive Opponent-wise Rollout Policy Optimization (AORPO). In AORPO, each agent builds its multi-agent environment model, consisting of a dynamics model and multiple opponent models,"},"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":"2105.03363","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-05-07T16:20:22Z","cross_cats_sorted":["cs.AI","cs.MA"],"title_canon_sha256":"6ce21e8e0ca46f9f1a3ffb822c81acceebb97c296d10bf40a57e73ec8a1d2188","abstract_canon_sha256":"ed91dd56cbc3334fa29f7673152305587825ae65719ffe0aafd2ec60dd21e8b4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:05:58.426207Z","signature_b64":"vtPFLVRAATsSbW1OwUyKj407PsCiWsVcT1NPbWdP+LsrlInCslFED5M9eor9H/QXYUDa5ri7nN2Z7w4tiCX5Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"26dfc7bd101f543a2421ee13ad6cbe5afac8542477690bc1555ddf458d87eb4e","last_reissued_at":"2026-07-05T04:05:58.425790Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:05:58.425790Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model-based Multi-agent Policy Optimization with Adaptive Opponent-wise Rollouts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.LG","authors_text":"Jian Shen, Ming Zhou, Weinan Zhang, Xihuai Wang","submitted_at":"2021-05-07T16:20:22Z","abstract_excerpt":"This paper investigates the model-based methods in multi-agent reinforcement learning (MARL). We specify the dynamics sample complexity and the opponent sample complexity in MARL, and conduct a theoretic analysis of return discrepancy upper bound. To reduce the upper bound with the intention of low sample complexity during the whole learning process, we propose a novel decentralized model-based MARL method, named Adaptive Opponent-wise Rollout Policy Optimization (AORPO). In AORPO, each agent builds its multi-agent environment model, consisting of a dynamics model and multiple opponent models,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.03363","kind":"arxiv","version":3},"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/2105.03363/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":"2105.03363","created_at":"2026-07-05T04:05:58.425846+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.03363v3","created_at":"2026-07-05T04:05:58.425846+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.03363","created_at":"2026-07-05T04:05:58.425846+00:00"},{"alias_kind":"pith_short_12","alias_value":"E3P4PPIQD5KD","created_at":"2026-07-05T04:05:58.425846+00:00"},{"alias_kind":"pith_short_16","alias_value":"E3P4PPIQD5KDUJBB","created_at":"2026-07-05T04:05:58.425846+00:00"},{"alias_kind":"pith_short_8","alias_value":"E3P4PPIQ","created_at":"2026-07-05T04:05:58.425846+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.23308","citing_title":"CODA: Coordination via On-Policy Diffusion for Multi-Agent Offline Reinforcement Learning","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E3P4PPIQD5KDUJBB5YJ223F6LL","json":"https://pith.science/pith/E3P4PPIQD5KDUJBB5YJ223F6LL.json","graph_json":"https://pith.science/api/pith-number/E3P4PPIQD5KDUJBB5YJ223F6LL/graph.json","events_json":"https://pith.science/api/pith-number/E3P4PPIQD5KDUJBB5YJ223F6LL/events.json","paper":"https://pith.science/paper/E3P4PPIQ"},"agent_actions":{"view_html":"https://pith.science/pith/E3P4PPIQD5KDUJBB5YJ223F6LL","download_json":"https://pith.science/pith/E3P4PPIQD5KDUJBB5YJ223F6LL.json","view_paper":"https://pith.science/paper/E3P4PPIQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.03363&json=true","fetch_graph":"https://pith.science/api/pith-number/E3P4PPIQD5KDUJBB5YJ223F6LL/graph.json","fetch_events":"https://pith.science/api/pith-number/E3P4PPIQD5KDUJBB5YJ223F6LL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E3P4PPIQD5KDUJBB5YJ223F6LL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E3P4PPIQD5KDUJBB5YJ223F6LL/action/storage_attestation","attest_author":"https://pith.science/pith/E3P4PPIQD5KDUJBB5YJ223F6LL/action/author_attestation","sign_citation":"https://pith.science/pith/E3P4PPIQD5KDUJBB5YJ223F6LL/action/citation_signature","submit_replication":"https://pith.science/pith/E3P4PPIQD5KDUJBB5YJ223F6LL/action/replication_record"}},"created_at":"2026-07-05T04:05:58.425846+00:00","updated_at":"2026-07-05T04:05:58.425846+00:00"}