{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KRIEML4UQFFAYLLU7Y67SZG5GE","short_pith_number":"pith:KRIEML4U","schema_version":"1.0","canonical_sha256":"5450462f94814a0c2d74fe3df964dd311eecb5465cbd54dec2510c7586c3db9c","source":{"kind":"arxiv","id":"2504.16415","version":1},"attestation_state":"computed","paper":{"title":"Natural Policy Gradient for Average Reward Non-Stationary RL","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Eshika Pathak, Gauri Joshi, Guannan Qu, Neharika Jali, Pranay Sharma","submitted_at":"2025-04-23T04:37:26Z","abstract_excerpt":"We consider the problem of non-stationary reinforcement learning (RL) in the infinite-horizon average-reward setting. We model it by a Markov Decision Process with time-varying rewards and transition probabilities, with a variation budget of $\\Delta_T$. Existing non-stationary RL algorithms focus on model-based and model-free value-based methods. Policy-based methods despite their flexibility in practice are not theoretically well understood in non-stationary RL. We propose and analyze the first model-free policy-based algorithm, Non-Stationary Natural Actor-Critic (NS-NAC), a policy gradient "},"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":"2504.16415","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-23T04:37:26Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"089d280c7659cef9009151ed6543b94242f089b45fc92fa704f8e51ba16f29d1","abstract_canon_sha256":"c977d7656a48777bb6c905289794dc48958981aacd4e54399661fd915d86a869"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:52:46.308070Z","signature_b64":"/Z33OCTKJmrJBBtPke8WbmSqcZm3GNn0NcTqUdNItAbsF+5hT7Mj4GG4E9gHe/3/2nLcWGyU6xZfFtEoiM27Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5450462f94814a0c2d74fe3df964dd311eecb5465cbd54dec2510c7586c3db9c","last_reissued_at":"2026-07-05T10:52:46.307648Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:52:46.307648Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Natural Policy Gradient for Average Reward Non-Stationary RL","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Eshika Pathak, Gauri Joshi, Guannan Qu, Neharika Jali, Pranay Sharma","submitted_at":"2025-04-23T04:37:26Z","abstract_excerpt":"We consider the problem of non-stationary reinforcement learning (RL) in the infinite-horizon average-reward setting. We model it by a Markov Decision Process with time-varying rewards and transition probabilities, with a variation budget of $\\Delta_T$. Existing non-stationary RL algorithms focus on model-based and model-free value-based methods. Policy-based methods despite their flexibility in practice are not theoretically well understood in non-stationary RL. We propose and analyze the first model-free policy-based algorithm, Non-Stationary Natural Actor-Critic (NS-NAC), a policy gradient "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.16415","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/2504.16415/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":"2504.16415","created_at":"2026-07-05T10:52:46.307706+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.16415v1","created_at":"2026-07-05T10:52:46.307706+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.16415","created_at":"2026-07-05T10:52:46.307706+00:00"},{"alias_kind":"pith_short_12","alias_value":"KRIEML4UQFFA","created_at":"2026-07-05T10:52:46.307706+00:00"},{"alias_kind":"pith_short_16","alias_value":"KRIEML4UQFFAYLLU","created_at":"2026-07-05T10:52:46.307706+00:00"},{"alias_kind":"pith_short_8","alias_value":"KRIEML4U","created_at":"2026-07-05T10:52:46.307706+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/KRIEML4UQFFAYLLU7Y67SZG5GE","json":"https://pith.science/pith/KRIEML4UQFFAYLLU7Y67SZG5GE.json","graph_json":"https://pith.science/api/pith-number/KRIEML4UQFFAYLLU7Y67SZG5GE/graph.json","events_json":"https://pith.science/api/pith-number/KRIEML4UQFFAYLLU7Y67SZG5GE/events.json","paper":"https://pith.science/paper/KRIEML4U"},"agent_actions":{"view_html":"https://pith.science/pith/KRIEML4UQFFAYLLU7Y67SZG5GE","download_json":"https://pith.science/pith/KRIEML4UQFFAYLLU7Y67SZG5GE.json","view_paper":"https://pith.science/paper/KRIEML4U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.16415&json=true","fetch_graph":"https://pith.science/api/pith-number/KRIEML4UQFFAYLLU7Y67SZG5GE/graph.json","fetch_events":"https://pith.science/api/pith-number/KRIEML4UQFFAYLLU7Y67SZG5GE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KRIEML4UQFFAYLLU7Y67SZG5GE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KRIEML4UQFFAYLLU7Y67SZG5GE/action/storage_attestation","attest_author":"https://pith.science/pith/KRIEML4UQFFAYLLU7Y67SZG5GE/action/author_attestation","sign_citation":"https://pith.science/pith/KRIEML4UQFFAYLLU7Y67SZG5GE/action/citation_signature","submit_replication":"https://pith.science/pith/KRIEML4UQFFAYLLU7Y67SZG5GE/action/replication_record"}},"created_at":"2026-07-05T10:52:46.307706+00:00","updated_at":"2026-07-05T10:52:46.307706+00:00"}