{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VAF2JTV2XZ56CYYFT5ZHC4YOPJ","short_pith_number":"pith:VAF2JTV2","schema_version":"1.0","canonical_sha256":"a80ba4cebabe7be163059f7271730e7a782668568bdd30ac91405ad970d76826","source":{"kind":"arxiv","id":"2607.28849","version":1},"attestation_state":"computed","paper":{"title":"Hypergradient-based Bilevel Reinforcement Learning with Improved Sample Complexity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Mudit Gaur, Naman Saxena, Vaneet Aggarwal","submitted_at":"2026-07-30T21:21:52Z","abstract_excerpt":"Bilevel reinforcement learning (RL) is an important framework within the literature of RL that can be used to formalize various categories of problems, such as meta-learning, hierarchical task decomposition, and reinforcement learning from human feedback (RL-HF). Most of the bilevel RL algorithms are either not scalable because of using hypergradient with Hessian, or they suffer from high sample complexity because of using penalty-based approximation methods. In this work, we propose a hypergradient-based bilevel RL algorithm using the optimality of the Boltzmann policy for the entropy regular"},"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":"2607.28849","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-30T21:21:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fe81af9897485e04ddc406293082784e4e70585c7b3efe81f89869f2bc3b44bb","abstract_canon_sha256":"fd6811b269eab6789ac12d79bd553ee6a199dc495f34e82a075cd660b9adc665"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-03T01:17:12.370260Z","signature_b64":"BKbtNxYFolkwny2WclQ8qUjigMebSx4WCh8TgG5c0nFIVqgFndvG/Vk6YXa1nhHRZISAkW6sGGugVgHMyYE1Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a80ba4cebabe7be163059f7271730e7a782668568bdd30ac91405ad970d76826","last_reissued_at":"2026-08-03T01:17:12.368674Z","signature_status":"signed_v1","first_computed_at":"2026-08-03T01:17:12.368674Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hypergradient-based Bilevel Reinforcement Learning with Improved Sample Complexity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Mudit Gaur, Naman Saxena, Vaneet Aggarwal","submitted_at":"2026-07-30T21:21:52Z","abstract_excerpt":"Bilevel reinforcement learning (RL) is an important framework within the literature of RL that can be used to formalize various categories of problems, such as meta-learning, hierarchical task decomposition, and reinforcement learning from human feedback (RL-HF). Most of the bilevel RL algorithms are either not scalable because of using hypergradient with Hessian, or they suffer from high sample complexity because of using penalty-based approximation methods. In this work, we propose a hypergradient-based bilevel RL algorithm using the optimality of the Boltzmann policy for the entropy regular"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.28849","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/2607.28849/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":"2607.28849","created_at":"2026-08-03T01:17:12.369639+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.28849v1","created_at":"2026-08-03T01:17:12.369639+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.28849","created_at":"2026-08-03T01:17:12.369639+00:00"},{"alias_kind":"pith_short_12","alias_value":"VAF2JTV2XZ56","created_at":"2026-08-03T01:17:12.369639+00:00"},{"alias_kind":"pith_short_16","alias_value":"VAF2JTV2XZ56CYYF","created_at":"2026-08-03T01:17:12.369639+00:00"},{"alias_kind":"pith_short_8","alias_value":"VAF2JTV2","created_at":"2026-08-03T01:17:12.369639+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VAF2JTV2XZ56CYYFT5ZHC4YOPJ","json":"https://pith.science/pith/VAF2JTV2XZ56CYYFT5ZHC4YOPJ.json","graph_json":"https://pith.science/api/pith-number/VAF2JTV2XZ56CYYFT5ZHC4YOPJ/graph.json","events_json":"https://pith.science/api/pith-number/VAF2JTV2XZ56CYYFT5ZHC4YOPJ/events.json","paper":"https://pith.science/paper/VAF2JTV2"},"agent_actions":{"view_html":"https://pith.science/pith/VAF2JTV2XZ56CYYFT5ZHC4YOPJ","download_json":"https://pith.science/pith/VAF2JTV2XZ56CYYFT5ZHC4YOPJ.json","view_paper":"https://pith.science/paper/VAF2JTV2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.28849&json=true","fetch_graph":"https://pith.science/api/pith-number/VAF2JTV2XZ56CYYFT5ZHC4YOPJ/graph.json","fetch_events":"https://pith.science/api/pith-number/VAF2JTV2XZ56CYYFT5ZHC4YOPJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VAF2JTV2XZ56CYYFT5ZHC4YOPJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VAF2JTV2XZ56CYYFT5ZHC4YOPJ/action/storage_attestation","attest_author":"https://pith.science/pith/VAF2JTV2XZ56CYYFT5ZHC4YOPJ/action/author_attestation","sign_citation":"https://pith.science/pith/VAF2JTV2XZ56CYYFT5ZHC4YOPJ/action/citation_signature","submit_replication":"https://pith.science/pith/VAF2JTV2XZ56CYYFT5ZHC4YOPJ/action/replication_record"}},"created_at":"2026-08-03T01:17:12.369639+00:00","updated_at":"2026-08-03T01:17:12.369639+00:00"}