{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:E246QXTCD74MBDFUQIYAIN5CGT","short_pith_number":"pith:E246QXTC","schema_version":"1.0","canonical_sha256":"26b9e85e621ff8c08cb482300437a234c0166906d9a2514b72d9e5788950c546","source":{"kind":"arxiv","id":"2411.12042","version":2},"attestation_state":"computed","paper":{"title":"Fast Convergence of Softmax Policy Mirror Ascent","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Issam Laradji, Nicolas Le Roux, Reza Asad, Reza Babanezhad, Sharan Vaswani","submitted_at":"2024-11-18T20:27:13Z","abstract_excerpt":"Natural policy gradient (NPG) is a common policy optimization algorithm and can be viewed as mirror ascent in the space of probabilities. Recently, Vaswani et al. [2021] introduced a policy gradient method that corresponds to mirror ascent in the dual space of logits. We refine this algorithm, removing its need for a normalization across actions and analyze the resulting method (referred to as SPMA). For tabular MDPs, we prove that SPMA with a constant step-size matches the linear convergence of NPG and achieves a faster convergence than constant step-size (accelerated) softmax 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":"2411.12042","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-11-18T20:27:13Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"c0f515bd9cf83b558c5b023166e2b209594d6a8b9c242c2aa8008e1d5ea22489","abstract_canon_sha256":"474ab19d9573630e505861fed161b017df65764da3ee30bdcb0938051faa522b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:22.349868Z","signature_b64":"/FsTj9Ao20i+0vvDfU88dSI4LuPlI2v+8Odmgi2eXLyWDtn58RB7HxKeD9BVwbH3+WRTfHUHaSp35vypWUbVBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"26b9e85e621ff8c08cb482300437a234c0166906d9a2514b72d9e5788950c546","last_reissued_at":"2026-07-05T11:12:22.349286Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:22.349286Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fast Convergence of Softmax Policy Mirror Ascent","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Issam Laradji, Nicolas Le Roux, Reza Asad, Reza Babanezhad, Sharan Vaswani","submitted_at":"2024-11-18T20:27:13Z","abstract_excerpt":"Natural policy gradient (NPG) is a common policy optimization algorithm and can be viewed as mirror ascent in the space of probabilities. Recently, Vaswani et al. [2021] introduced a policy gradient method that corresponds to mirror ascent in the dual space of logits. We refine this algorithm, removing its need for a normalization across actions and analyze the resulting method (referred to as SPMA). For tabular MDPs, we prove that SPMA with a constant step-size matches the linear convergence of NPG and achieves a faster convergence than constant step-size (accelerated) softmax policy gradient"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.12042","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/2411.12042/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":"2411.12042","created_at":"2026-07-05T11:12:22.349354+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.12042v2","created_at":"2026-07-05T11:12:22.349354+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.12042","created_at":"2026-07-05T11:12:22.349354+00:00"},{"alias_kind":"pith_short_12","alias_value":"E246QXTCD74M","created_at":"2026-07-05T11:12:22.349354+00:00"},{"alias_kind":"pith_short_16","alias_value":"E246QXTCD74MBDFU","created_at":"2026-07-05T11:12:22.349354+00:00"},{"alias_kind":"pith_short_8","alias_value":"E246QXTC","created_at":"2026-07-05T11:12:22.349354+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.09838","citing_title":"Dissecting Discrete Soft Actor-Critic: Limitations and Principled Alternatives","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E246QXTCD74MBDFUQIYAIN5CGT","json":"https://pith.science/pith/E246QXTCD74MBDFUQIYAIN5CGT.json","graph_json":"https://pith.science/api/pith-number/E246QXTCD74MBDFUQIYAIN5CGT/graph.json","events_json":"https://pith.science/api/pith-number/E246QXTCD74MBDFUQIYAIN5CGT/events.json","paper":"https://pith.science/paper/E246QXTC"},"agent_actions":{"view_html":"https://pith.science/pith/E246QXTCD74MBDFUQIYAIN5CGT","download_json":"https://pith.science/pith/E246QXTCD74MBDFUQIYAIN5CGT.json","view_paper":"https://pith.science/paper/E246QXTC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.12042&json=true","fetch_graph":"https://pith.science/api/pith-number/E246QXTCD74MBDFUQIYAIN5CGT/graph.json","fetch_events":"https://pith.science/api/pith-number/E246QXTCD74MBDFUQIYAIN5CGT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E246QXTCD74MBDFUQIYAIN5CGT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E246QXTCD74MBDFUQIYAIN5CGT/action/storage_attestation","attest_author":"https://pith.science/pith/E246QXTCD74MBDFUQIYAIN5CGT/action/author_attestation","sign_citation":"https://pith.science/pith/E246QXTCD74MBDFUQIYAIN5CGT/action/citation_signature","submit_replication":"https://pith.science/pith/E246QXTCD74MBDFUQIYAIN5CGT/action/replication_record"}},"created_at":"2026-07-05T11:12:22.349354+00:00","updated_at":"2026-07-05T11:12:22.349354+00:00"}