{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:7P2HD4LJJ2PENE3I2TSPXYWSY5","short_pith_number":"pith:7P2HD4LJ","schema_version":"1.0","canonical_sha256":"fbf471f1694e9e469368d4e4fbe2d2c751bdf4b74c2a59c642fa72d971ad1132","source":{"kind":"arxiv","id":"2009.14180","version":2},"attestation_state":"computed","paper":{"title":"Learning to Play against Any Mixture of Opponents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.MA","authors_text":"Max Olan Smith, Michael P. Wellman, Thomas Anthony, Yongzhao Wang","submitted_at":"2020-09-29T17:48:10Z","abstract_excerpt":"Intuitively, experience playing against one mixture of opponents in a given domain should be relevant for a different mixture in the same domain. We propose a transfer learning method, Q-Mixing, that starts by learning Q-values against each pure-strategy opponent. Then a Q-value for any distribution of opponent strategies is approximated by appropriately averaging the separately learned Q-values. From these components, we construct policies against all opponent mixtures without any further training. We empirically validate Q-Mixing in two environments: a simple grid-world soccer environment, a"},"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":"2009.14180","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MA","submitted_at":"2020-09-29T17:48:10Z","cross_cats_sorted":[],"title_canon_sha256":"4fe84b84277b9a70d21ca7d02c0004e698c60b45d6e2b8690278795c0ed1e604","abstract_canon_sha256":"497493c53dd95caa04bb434005c21aeab7117150ef4390302c9361a8bb105df3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:46:00.891854Z","signature_b64":"5FVoDXwhOmqUMkC3f+S1Pk3xvUhD3J0vqOrvirjIa7Q6nuqfImIvz3o32rqHMVaxrd8WMaEpmqSlkhsHnpQRCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fbf471f1694e9e469368d4e4fbe2d2c751bdf4b74c2a59c642fa72d971ad1132","last_reissued_at":"2026-07-05T02:46:00.891416Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:46:00.891416Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Play against Any Mixture of Opponents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.MA","authors_text":"Max Olan Smith, Michael P. Wellman, Thomas Anthony, Yongzhao Wang","submitted_at":"2020-09-29T17:48:10Z","abstract_excerpt":"Intuitively, experience playing against one mixture of opponents in a given domain should be relevant for a different mixture in the same domain. We propose a transfer learning method, Q-Mixing, that starts by learning Q-values against each pure-strategy opponent. Then a Q-value for any distribution of opponent strategies is approximated by appropriately averaging the separately learned Q-values. From these components, we construct policies against all opponent mixtures without any further training. We empirically validate Q-Mixing in two environments: a simple grid-world soccer environment, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2009.14180","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/2009.14180/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":"2009.14180","created_at":"2026-07-05T02:46:00.891474+00:00"},{"alias_kind":"arxiv_version","alias_value":"2009.14180v2","created_at":"2026-07-05T02:46:00.891474+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2009.14180","created_at":"2026-07-05T02:46:00.891474+00:00"},{"alias_kind":"pith_short_12","alias_value":"7P2HD4LJJ2PE","created_at":"2026-07-05T02:46:00.891474+00:00"},{"alias_kind":"pith_short_16","alias_value":"7P2HD4LJJ2PENE3I","created_at":"2026-07-05T02:46:00.891474+00:00"},{"alias_kind":"pith_short_8","alias_value":"7P2HD4LJ","created_at":"2026-07-05T02:46:00.891474+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.19488","citing_title":"PoolFlip: A Multi-Agent Reinforcement Learning Security Environment for Cyber Defense","ref_index":29,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7P2HD4LJJ2PENE3I2TSPXYWSY5","json":"https://pith.science/pith/7P2HD4LJJ2PENE3I2TSPXYWSY5.json","graph_json":"https://pith.science/api/pith-number/7P2HD4LJJ2PENE3I2TSPXYWSY5/graph.json","events_json":"https://pith.science/api/pith-number/7P2HD4LJJ2PENE3I2TSPXYWSY5/events.json","paper":"https://pith.science/paper/7P2HD4LJ"},"agent_actions":{"view_html":"https://pith.science/pith/7P2HD4LJJ2PENE3I2TSPXYWSY5","download_json":"https://pith.science/pith/7P2HD4LJJ2PENE3I2TSPXYWSY5.json","view_paper":"https://pith.science/paper/7P2HD4LJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2009.14180&json=true","fetch_graph":"https://pith.science/api/pith-number/7P2HD4LJJ2PENE3I2TSPXYWSY5/graph.json","fetch_events":"https://pith.science/api/pith-number/7P2HD4LJJ2PENE3I2TSPXYWSY5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7P2HD4LJJ2PENE3I2TSPXYWSY5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7P2HD4LJJ2PENE3I2TSPXYWSY5/action/storage_attestation","attest_author":"https://pith.science/pith/7P2HD4LJJ2PENE3I2TSPXYWSY5/action/author_attestation","sign_citation":"https://pith.science/pith/7P2HD4LJJ2PENE3I2TSPXYWSY5/action/citation_signature","submit_replication":"https://pith.science/pith/7P2HD4LJJ2PENE3I2TSPXYWSY5/action/replication_record"}},"created_at":"2026-07-05T02:46:00.891474+00:00","updated_at":"2026-07-05T02:46:00.891474+00:00"}