{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:7C566PZGX2M27CWSZNT3FJG4XL","short_pith_number":"pith:7C566PZG","schema_version":"1.0","canonical_sha256":"f8bbef3f26be99af8ad2cb67b2a4dcbad20e6b6dca64f30b32902c7718844757","source":{"kind":"arxiv","id":"2205.01447","version":2},"attestation_state":"computed","paper":{"title":"Model-Free Opponent Shaping","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.AI","authors_text":"Chris Lu, Christian Schroeder de Witt, Jakob Foerster, Timon Willi","submitted_at":"2022-05-03T12:20:14Z","abstract_excerpt":"In general-sum games, the interaction of self-interested learning agents commonly leads to collectively worst-case outcomes, such as defect-defect in the iterated prisoner's dilemma (IPD). To overcome this, some methods, such as Learning with Opponent-Learning Awareness (LOLA), shape their opponents' learning process. However, these methods are myopic since only a small number of steps can be anticipated, are asymmetric since they treat other agents as naive learners, and require the use of higher-order derivatives, which are calculated through white-box access to an opponent's differentiable "},"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":"2205.01447","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2022-05-03T12:20:14Z","cross_cats_sorted":["cs.MA"],"title_canon_sha256":"aac8cea3d5a6acd6f4d1e94610edef134235b1511b6a666f08f2dbc5a7fe37cd","abstract_canon_sha256":"9529efdff60dc179f21d3a121589387f53cfa17c3d23e35def5bf3e25c250f75"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:13:06.709784Z","signature_b64":"WcUV/Gbhs/GZVzsUeslNVJmwzkkab7k0wu3D8dsy6Ct+ftrFTvvVYEa+LVmEnj4V3jtZeSOgsL51mpBtn5qyAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f8bbef3f26be99af8ad2cb67b2a4dcbad20e6b6dca64f30b32902c7718844757","last_reissued_at":"2026-07-05T05:13:06.709353Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:13:06.709353Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model-Free Opponent Shaping","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.AI","authors_text":"Chris Lu, Christian Schroeder de Witt, Jakob Foerster, Timon Willi","submitted_at":"2022-05-03T12:20:14Z","abstract_excerpt":"In general-sum games, the interaction of self-interested learning agents commonly leads to collectively worst-case outcomes, such as defect-defect in the iterated prisoner's dilemma (IPD). To overcome this, some methods, such as Learning with Opponent-Learning Awareness (LOLA), shape their opponents' learning process. However, these methods are myopic since only a small number of steps can be anticipated, are asymmetric since they treat other agents as naive learners, and require the use of higher-order derivatives, which are calculated through white-box access to an opponent's differentiable "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.01447","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/2205.01447/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":"2205.01447","created_at":"2026-07-05T05:13:06.709413+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.01447v2","created_at":"2026-07-05T05:13:06.709413+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.01447","created_at":"2026-07-05T05:13:06.709413+00:00"},{"alias_kind":"pith_short_12","alias_value":"7C566PZGX2M2","created_at":"2026-07-05T05:13:06.709413+00:00"},{"alias_kind":"pith_short_16","alias_value":"7C566PZGX2M27CWS","created_at":"2026-07-05T05:13:06.709413+00:00"},{"alias_kind":"pith_short_8","alias_value":"7C566PZG","created_at":"2026-07-05T05:13:06.709413+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.02292","citing_title":"One More Time: Revisiting Neural Quantum States from a Reinforcement Learning Perspective","ref_index":127,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18123","citing_title":"ConventionPlay: Capability-Limited Training for Robust Ad-Hoc Collaboration","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7C566PZGX2M27CWSZNT3FJG4XL","json":"https://pith.science/pith/7C566PZGX2M27CWSZNT3FJG4XL.json","graph_json":"https://pith.science/api/pith-number/7C566PZGX2M27CWSZNT3FJG4XL/graph.json","events_json":"https://pith.science/api/pith-number/7C566PZGX2M27CWSZNT3FJG4XL/events.json","paper":"https://pith.science/paper/7C566PZG"},"agent_actions":{"view_html":"https://pith.science/pith/7C566PZGX2M27CWSZNT3FJG4XL","download_json":"https://pith.science/pith/7C566PZGX2M27CWSZNT3FJG4XL.json","view_paper":"https://pith.science/paper/7C566PZG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.01447&json=true","fetch_graph":"https://pith.science/api/pith-number/7C566PZGX2M27CWSZNT3FJG4XL/graph.json","fetch_events":"https://pith.science/api/pith-number/7C566PZGX2M27CWSZNT3FJG4XL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7C566PZGX2M27CWSZNT3FJG4XL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7C566PZGX2M27CWSZNT3FJG4XL/action/storage_attestation","attest_author":"https://pith.science/pith/7C566PZGX2M27CWSZNT3FJG4XL/action/author_attestation","sign_citation":"https://pith.science/pith/7C566PZGX2M27CWSZNT3FJG4XL/action/citation_signature","submit_replication":"https://pith.science/pith/7C566PZGX2M27CWSZNT3FJG4XL/action/replication_record"}},"created_at":"2026-07-05T05:13:06.709413+00:00","updated_at":"2026-07-05T05:13:06.709413+00:00"}