{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:NLRW4SCCLTWY5JDNBNOA5HQMNP","short_pith_number":"pith:NLRW4SCC","schema_version":"1.0","canonical_sha256":"6ae36e48425ced8ea46d0b5c0e9e0c6bd75ea56f894743a985bd59ddbbbd1563","source":{"kind":"arxiv","id":"2212.02733","version":3},"attestation_state":"computed","paper":{"title":"CURO: Curriculum Learning for Relative Overgeneralization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.LG","authors_text":"Bei Peng, Lin Shi, Qiyuan Liu","submitted_at":"2022-12-06T03:41:08Z","abstract_excerpt":"Relative overgeneralization (RO) is a pathology that can arise in cooperative multi-agent tasks when the optimal joint action's utility falls below that of a sub-optimal joint action. RO can cause the agents to get stuck into local optima or fail to solve cooperative tasks requiring significant coordination between agents within a given timestep. In this work, we empirically find that, in multi-agent reinforcement learning (MARL), both value-based and policy gradient MARL algorithms can suffer from RO and fail to learn effective coordination policies. To better overcome RO, we propose a novel "},"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":"2212.02733","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-12-06T03:41:08Z","cross_cats_sorted":["cs.AI","cs.MA"],"title_canon_sha256":"cb07ee540e6da9fa96f7e808d19c8bda952f2591172809aff3fc578a17e1955b","abstract_canon_sha256":"4929f007a1fbd30713bbdf7f44cecf02d997b8c7c0f0466d3e94eca29a4d522a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:10:28.779982Z","signature_b64":"BWPu5CGSa+igaTDjkWylfwkMolBERs3XXqPUxpYX+xSnttS86IccHh1WnriDeMPBMTgIxx1Fj4a511/fQdfeCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ae36e48425ced8ea46d0b5c0e9e0c6bd75ea56f894743a985bd59ddbbbd1563","last_reissued_at":"2026-07-05T09:10:28.779628Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:10:28.779628Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CURO: Curriculum Learning for Relative Overgeneralization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.LG","authors_text":"Bei Peng, Lin Shi, Qiyuan Liu","submitted_at":"2022-12-06T03:41:08Z","abstract_excerpt":"Relative overgeneralization (RO) is a pathology that can arise in cooperative multi-agent tasks when the optimal joint action's utility falls below that of a sub-optimal joint action. RO can cause the agents to get stuck into local optima or fail to solve cooperative tasks requiring significant coordination between agents within a given timestep. In this work, we empirically find that, in multi-agent reinforcement learning (MARL), both value-based and policy gradient MARL algorithms can suffer from RO and fail to learn effective coordination policies. To better overcome RO, we propose a novel "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.02733","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/2212.02733/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":"2212.02733","created_at":"2026-07-05T09:10:28.779683+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.02733v3","created_at":"2026-07-05T09:10:28.779683+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.02733","created_at":"2026-07-05T09:10:28.779683+00:00"},{"alias_kind":"pith_short_12","alias_value":"NLRW4SCCLTWY","created_at":"2026-07-05T09:10:28.779683+00:00"},{"alias_kind":"pith_short_16","alias_value":"NLRW4SCCLTWY5JDN","created_at":"2026-07-05T09:10:28.779683+00:00"},{"alias_kind":"pith_short_8","alias_value":"NLRW4SCC","created_at":"2026-07-05T09:10:28.779683+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.11099","citing_title":"Mitigating Relative Over-Generalization in Multi-Agent Reinforcement Learning","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NLRW4SCCLTWY5JDNBNOA5HQMNP","json":"https://pith.science/pith/NLRW4SCCLTWY5JDNBNOA5HQMNP.json","graph_json":"https://pith.science/api/pith-number/NLRW4SCCLTWY5JDNBNOA5HQMNP/graph.json","events_json":"https://pith.science/api/pith-number/NLRW4SCCLTWY5JDNBNOA5HQMNP/events.json","paper":"https://pith.science/paper/NLRW4SCC"},"agent_actions":{"view_html":"https://pith.science/pith/NLRW4SCCLTWY5JDNBNOA5HQMNP","download_json":"https://pith.science/pith/NLRW4SCCLTWY5JDNBNOA5HQMNP.json","view_paper":"https://pith.science/paper/NLRW4SCC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.02733&json=true","fetch_graph":"https://pith.science/api/pith-number/NLRW4SCCLTWY5JDNBNOA5HQMNP/graph.json","fetch_events":"https://pith.science/api/pith-number/NLRW4SCCLTWY5JDNBNOA5HQMNP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NLRW4SCCLTWY5JDNBNOA5HQMNP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NLRW4SCCLTWY5JDNBNOA5HQMNP/action/storage_attestation","attest_author":"https://pith.science/pith/NLRW4SCCLTWY5JDNBNOA5HQMNP/action/author_attestation","sign_citation":"https://pith.science/pith/NLRW4SCCLTWY5JDNBNOA5HQMNP/action/citation_signature","submit_replication":"https://pith.science/pith/NLRW4SCCLTWY5JDNBNOA5HQMNP/action/replication_record"}},"created_at":"2026-07-05T09:10:28.779683+00:00","updated_at":"2026-07-05T09:10:28.779683+00:00"}