{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:IPVEHNFDKGIITZTV72HWCKOPT7","short_pith_number":"pith:IPVEHNFD","schema_version":"1.0","canonical_sha256":"43ea43b4a3519089e675fe8f6129cf9fe4b8362f25128b23af17ccc983e83279","source":{"kind":"arxiv","id":"2001.01866","version":2},"attestation_state":"computed","paper":{"title":"Reinforcement Learning via Fenchel-Rockafellar Duality","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Bo Dai, Ofir Nachum","submitted_at":"2020-01-07T02:59:59Z","abstract_excerpt":"We review basic concepts of convex duality, focusing on the very general and supremely useful Fenchel-Rockafellar duality. We summarize how this duality may be applied to a variety of reinforcement learning (RL) settings, including policy evaluation or optimization, online or offline learning, and discounted or undiscounted rewards. The derivations yield a number of intriguing results, including the ability to perform policy evaluation and on-policy policy gradient with behavior-agnostic offline data and methods to learn a policy via max-likelihood optimization. Although many of these results "},"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":"2001.01866","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-01-07T02:59:59Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"0c46dba128aca11da70ce113a2d9af875d04cd7d9a2558d7cb76d3e9dd9fb8ab","abstract_canon_sha256":"3133818af109e1d37fee6f9f5175f48282c82883fb8ed917fee124a996152dcd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:32:40.800135Z","signature_b64":"ubwOKuEcZD7EIANlHOiecGpIgankshG+Dz5dbinnhdRzLWHAB89u/D6MPcCsfm7TFykAjBLixH/1fpddTkIjAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"43ea43b4a3519089e675fe8f6129cf9fe4b8362f25128b23af17ccc983e83279","last_reissued_at":"2026-07-05T00:32:40.799710Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:32:40.799710Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reinforcement Learning via Fenchel-Rockafellar Duality","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Bo Dai, Ofir Nachum","submitted_at":"2020-01-07T02:59:59Z","abstract_excerpt":"We review basic concepts of convex duality, focusing on the very general and supremely useful Fenchel-Rockafellar duality. We summarize how this duality may be applied to a variety of reinforcement learning (RL) settings, including policy evaluation or optimization, online or offline learning, and discounted or undiscounted rewards. The derivations yield a number of intriguing results, including the ability to perform policy evaluation and on-policy policy gradient with behavior-agnostic offline data and methods to learn a policy via max-likelihood optimization. Although many of these results "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2001.01866","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/2001.01866/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":"2001.01866","created_at":"2026-07-05T00:32:40.799773+00:00"},{"alias_kind":"arxiv_version","alias_value":"2001.01866v2","created_at":"2026-07-05T00:32:40.799773+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2001.01866","created_at":"2026-07-05T00:32:40.799773+00:00"},{"alias_kind":"pith_short_12","alias_value":"IPVEHNFDKGII","created_at":"2026-07-05T00:32:40.799773+00:00"},{"alias_kind":"pith_short_16","alias_value":"IPVEHNFDKGIITZTV","created_at":"2026-07-05T00:32:40.799773+00:00"},{"alias_kind":"pith_short_8","alias_value":"IPVEHNFD","created_at":"2026-07-05T00:32:40.799773+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2408.08812","citing_title":"TRAM: Test-Time Risk Adaptation with Mixture of Agents","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2408.16286","citing_title":"Near-Optimal Policy Identification in Robust Constrained Markov Decision Processes via Epigraph Form","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2509.15103","citing_title":"Vulnerable Agent Identification in Large-Scale Multi-Agent Reinforcement Learning","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2005.01643","citing_title":"Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems","ref_index":137,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IPVEHNFDKGIITZTV72HWCKOPT7","json":"https://pith.science/pith/IPVEHNFDKGIITZTV72HWCKOPT7.json","graph_json":"https://pith.science/api/pith-number/IPVEHNFDKGIITZTV72HWCKOPT7/graph.json","events_json":"https://pith.science/api/pith-number/IPVEHNFDKGIITZTV72HWCKOPT7/events.json","paper":"https://pith.science/paper/IPVEHNFD"},"agent_actions":{"view_html":"https://pith.science/pith/IPVEHNFDKGIITZTV72HWCKOPT7","download_json":"https://pith.science/pith/IPVEHNFDKGIITZTV72HWCKOPT7.json","view_paper":"https://pith.science/paper/IPVEHNFD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2001.01866&json=true","fetch_graph":"https://pith.science/api/pith-number/IPVEHNFDKGIITZTV72HWCKOPT7/graph.json","fetch_events":"https://pith.science/api/pith-number/IPVEHNFDKGIITZTV72HWCKOPT7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IPVEHNFDKGIITZTV72HWCKOPT7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IPVEHNFDKGIITZTV72HWCKOPT7/action/storage_attestation","attest_author":"https://pith.science/pith/IPVEHNFDKGIITZTV72HWCKOPT7/action/author_attestation","sign_citation":"https://pith.science/pith/IPVEHNFDKGIITZTV72HWCKOPT7/action/citation_signature","submit_replication":"https://pith.science/pith/IPVEHNFDKGIITZTV72HWCKOPT7/action/replication_record"}},"created_at":"2026-07-05T00:32:40.799773+00:00","updated_at":"2026-07-05T00:32:40.799773+00:00"}