{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:BO3ZKCSO6QXIVKNADL2QQINSMR","short_pith_number":"pith:BO3ZKCSO","schema_version":"1.0","canonical_sha256":"0bb7950a4ef42e8aa9a01af50821b2646ff78e93084547b939ad0d03c7195762","source":{"kind":"arxiv","id":"2209.10072","version":1},"attestation_state":"computed","paper":{"title":"On the Convergence Theory of Meta Reinforcement Learning with Personalized Policies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Dong Li, Haozhi Wang, Jianye Hao, Qing Wang, Yinchuan Li, Yunfeng Shao","submitted_at":"2022-09-21T02:27:56Z","abstract_excerpt":"Modern meta-reinforcement learning (Meta-RL) methods are mainly developed based on model-agnostic meta-learning, which performs policy gradient steps across tasks to maximize policy performance. However, the gradient conflict problem is still poorly understood in Meta-RL, which may lead to performance degradation when encountering distinct tasks. To tackle this challenge, this paper proposes a novel personalized Meta-RL (pMeta-RL) algorithm, which aggregates task-specific personalized policies to update a meta-policy used for all tasks, while maintaining personalized policies to maximize the 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":"2209.10072","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2022-09-21T02:27:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"5a52d0a5c476f056cf70d3860246013b53b36a2b71a16889a1779c9df97827c2","abstract_canon_sha256":"40f34996429bae9c1c15d635e99df71db19af0b5f12150fffb8994a8635341c1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:59:44.719881Z","signature_b64":"byaMTteUbhFH/Sr918jBflTLTIZfsPS6TyxQjPzMX/jnAbybYV36ficY41bv57FaFg/gBmNzcEYzYns3MhE1AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0bb7950a4ef42e8aa9a01af50821b2646ff78e93084547b939ad0d03c7195762","last_reissued_at":"2026-07-05T04:59:44.719404Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:59:44.719404Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Convergence Theory of Meta Reinforcement Learning with Personalized Policies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Dong Li, Haozhi Wang, Jianye Hao, Qing Wang, Yinchuan Li, Yunfeng Shao","submitted_at":"2022-09-21T02:27:56Z","abstract_excerpt":"Modern meta-reinforcement learning (Meta-RL) methods are mainly developed based on model-agnostic meta-learning, which performs policy gradient steps across tasks to maximize policy performance. However, the gradient conflict problem is still poorly understood in Meta-RL, which may lead to performance degradation when encountering distinct tasks. To tackle this challenge, this paper proposes a novel personalized Meta-RL (pMeta-RL) algorithm, which aggregates task-specific personalized policies to update a meta-policy used for all tasks, while maintaining personalized policies to maximize the a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.10072","kind":"arxiv","version":1},"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/2209.10072/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":"2209.10072","created_at":"2026-07-05T04:59:44.719462+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.10072v1","created_at":"2026-07-05T04:59:44.719462+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.10072","created_at":"2026-07-05T04:59:44.719462+00:00"},{"alias_kind":"pith_short_12","alias_value":"BO3ZKCSO6QXI","created_at":"2026-07-05T04:59:44.719462+00:00"},{"alias_kind":"pith_short_16","alias_value":"BO3ZKCSO6QXIVKNA","created_at":"2026-07-05T04:59:44.719462+00:00"},{"alias_kind":"pith_short_8","alias_value":"BO3ZKCSO","created_at":"2026-07-05T04:59:44.719462+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.17036","citing_title":"LiveGraph: Active-Structure Neural Re-ranking for Exercise Recommendation","ref_index":52,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BO3ZKCSO6QXIVKNADL2QQINSMR","json":"https://pith.science/pith/BO3ZKCSO6QXIVKNADL2QQINSMR.json","graph_json":"https://pith.science/api/pith-number/BO3ZKCSO6QXIVKNADL2QQINSMR/graph.json","events_json":"https://pith.science/api/pith-number/BO3ZKCSO6QXIVKNADL2QQINSMR/events.json","paper":"https://pith.science/paper/BO3ZKCSO"},"agent_actions":{"view_html":"https://pith.science/pith/BO3ZKCSO6QXIVKNADL2QQINSMR","download_json":"https://pith.science/pith/BO3ZKCSO6QXIVKNADL2QQINSMR.json","view_paper":"https://pith.science/paper/BO3ZKCSO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.10072&json=true","fetch_graph":"https://pith.science/api/pith-number/BO3ZKCSO6QXIVKNADL2QQINSMR/graph.json","fetch_events":"https://pith.science/api/pith-number/BO3ZKCSO6QXIVKNADL2QQINSMR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BO3ZKCSO6QXIVKNADL2QQINSMR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BO3ZKCSO6QXIVKNADL2QQINSMR/action/storage_attestation","attest_author":"https://pith.science/pith/BO3ZKCSO6QXIVKNADL2QQINSMR/action/author_attestation","sign_citation":"https://pith.science/pith/BO3ZKCSO6QXIVKNADL2QQINSMR/action/citation_signature","submit_replication":"https://pith.science/pith/BO3ZKCSO6QXIVKNADL2QQINSMR/action/replication_record"}},"created_at":"2026-07-05T04:59:44.719462+00:00","updated_at":"2026-07-05T04:59:44.719462+00:00"}