{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:HPUR5IGCDAP5BRGTJP4NIELTVV","short_pith_number":"pith:HPUR5IGC","schema_version":"1.0","canonical_sha256":"3be91ea0c2181fd0c4d34bf8d41173ad5d66aa7a47b0d71cc6faaf09fdd7209e","source":{"kind":"arxiv","id":"2109.11867","version":2},"attestation_state":"computed","paper":{"title":"The $f$-Divergence Reinforcement Learning Framework","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chen Gong, Guoliang Fan, Qiang He, Xianjie Zhang, Xiaoyu Chen, Xinwen Hou, Yu Liu, Yunpeng Bai, Zhou Yang","submitted_at":"2021-09-24T10:20:46Z","abstract_excerpt":"The framework of deep reinforcement learning (DRL) provides a powerful and widely applicable mathematical formalization for sequential decision-making. This paper present a novel DRL framework, termed \\emph{$f$-Divergence Reinforcement Learning (FRL)}. In FRL, the policy evaluation and policy improvement phases are simultaneously performed by minimizing the $f$-divergence between the learning policy and sampling policy, which is distinct from conventional DRL algorithms that aim to maximize the expected cumulative rewards. We theoretically prove that minimizing such $f$-divergence can make the"},"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":"2109.11867","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-09-24T10:20:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8ecf9aaa8a022184763eb9311c1b9094c280bdf4d69a8196712f4ad86f390616","abstract_canon_sha256":"bf9eca90f32ac1654b6e519f80e98604e4d81004063d2e149776eb152dbe3b9b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:40:47.509821Z","signature_b64":"PqZFTI6c7AAiKvcldG8CSII9M0IxTBceg6PKXt00OtN5kSdw5BldFMH2NijbDbvb8Ondj8TDYiUxM8NHKAIyAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3be91ea0c2181fd0c4d34bf8d41173ad5d66aa7a47b0d71cc6faaf09fdd7209e","last_reissued_at":"2026-07-05T03:40:47.509328Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:40:47.509328Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The $f$-Divergence Reinforcement Learning Framework","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chen Gong, Guoliang Fan, Qiang He, Xianjie Zhang, Xiaoyu Chen, Xinwen Hou, Yu Liu, Yunpeng Bai, Zhou Yang","submitted_at":"2021-09-24T10:20:46Z","abstract_excerpt":"The framework of deep reinforcement learning (DRL) provides a powerful and widely applicable mathematical formalization for sequential decision-making. This paper present a novel DRL framework, termed \\emph{$f$-Divergence Reinforcement Learning (FRL)}. In FRL, the policy evaluation and policy improvement phases are simultaneously performed by minimizing the $f$-divergence between the learning policy and sampling policy, which is distinct from conventional DRL algorithms that aim to maximize the expected cumulative rewards. We theoretically prove that minimizing such $f$-divergence can make the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.11867","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/2109.11867/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":"2109.11867","created_at":"2026-07-05T03:40:47.509396+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.11867v2","created_at":"2026-07-05T03:40:47.509396+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.11867","created_at":"2026-07-05T03:40:47.509396+00:00"},{"alias_kind":"pith_short_12","alias_value":"HPUR5IGCDAP5","created_at":"2026-07-05T03:40:47.509396+00:00"},{"alias_kind":"pith_short_16","alias_value":"HPUR5IGCDAP5BRGT","created_at":"2026-07-05T03:40:47.509396+00:00"},{"alias_kind":"pith_short_8","alias_value":"HPUR5IGC","created_at":"2026-07-05T03:40:47.509396+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.18768","citing_title":"Diversity By Design: Leveraging Distribution Matching for Offline Model-Based Optimization","ref_index":2021,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HPUR5IGCDAP5BRGTJP4NIELTVV","json":"https://pith.science/pith/HPUR5IGCDAP5BRGTJP4NIELTVV.json","graph_json":"https://pith.science/api/pith-number/HPUR5IGCDAP5BRGTJP4NIELTVV/graph.json","events_json":"https://pith.science/api/pith-number/HPUR5IGCDAP5BRGTJP4NIELTVV/events.json","paper":"https://pith.science/paper/HPUR5IGC"},"agent_actions":{"view_html":"https://pith.science/pith/HPUR5IGCDAP5BRGTJP4NIELTVV","download_json":"https://pith.science/pith/HPUR5IGCDAP5BRGTJP4NIELTVV.json","view_paper":"https://pith.science/paper/HPUR5IGC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.11867&json=true","fetch_graph":"https://pith.science/api/pith-number/HPUR5IGCDAP5BRGTJP4NIELTVV/graph.json","fetch_events":"https://pith.science/api/pith-number/HPUR5IGCDAP5BRGTJP4NIELTVV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HPUR5IGCDAP5BRGTJP4NIELTVV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HPUR5IGCDAP5BRGTJP4NIELTVV/action/storage_attestation","attest_author":"https://pith.science/pith/HPUR5IGCDAP5BRGTJP4NIELTVV/action/author_attestation","sign_citation":"https://pith.science/pith/HPUR5IGCDAP5BRGTJP4NIELTVV/action/citation_signature","submit_replication":"https://pith.science/pith/HPUR5IGCDAP5BRGTJP4NIELTVV/action/replication_record"}},"created_at":"2026-07-05T03:40:47.509396+00:00","updated_at":"2026-07-05T03:40:47.509396+00:00"}