{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZVXQMPBXUKAGDRDINYPSZTDIBZ","short_pith_number":"pith:ZVXQMPBX","schema_version":"1.0","canonical_sha256":"cd6f063c37a28061c4686e1f2ccc680e736c8cb86fc1fa45af0a6aceffd6cf98","source":{"kind":"arxiv","id":"2605.24357","version":1},"attestation_state":"computed","paper":{"title":"Refined Analysis of Entropy-Regularized Actor-Critic","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Daniil Tiapkin, Eric Moulines, Paul Mangold, Safwan Labbi","submitted_at":"2026-05-23T02:41:07Z","abstract_excerpt":"In this paper, we study the role of the critic in actor--critic for entropy-regularized, finite, discounted environments. We establish that, when the critic is exact, using the latter as a baseline is a variance-reduction method in a strong sense. In this case, actor--critic with stochastic gradients matches the sample complexity of deterministic policy gradient, reaching an $\\epsilon$-optimal regularized value with $\\tilde{O}(\\log(1/\\epsilon))$ samples. In practice, the critic is learned alongside the actor: the variance of the actor update is then influenced by the critic's variance and bias"},"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":"2605.24357","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T02:41:07Z","cross_cats_sorted":[],"title_canon_sha256":"5fffcfa5e1a3c34d06a123424309c9578e7ccebc9441ec8827862cb7555c132a","abstract_canon_sha256":"31d67085d3cec2e5bf0553559dd5f7d9ab886474b31a4fcb91c88bcfd0b737ab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:01.669101Z","signature_b64":"jP4jA4hOLDhybAdemVqaHQ2SLVs44+JcoFWd7aqf8J3AbhuxFsg6jOdONeNjSWOhq1T9/5B1P2qIn+mh194DCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cd6f063c37a28061c4686e1f2ccc680e736c8cb86fc1fa45af0a6aceffd6cf98","last_reissued_at":"2026-05-26T01:03:01.668505Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:01.668505Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Refined Analysis of Entropy-Regularized Actor-Critic","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Daniil Tiapkin, Eric Moulines, Paul Mangold, Safwan Labbi","submitted_at":"2026-05-23T02:41:07Z","abstract_excerpt":"In this paper, we study the role of the critic in actor--critic for entropy-regularized, finite, discounted environments. We establish that, when the critic is exact, using the latter as a baseline is a variance-reduction method in a strong sense. In this case, actor--critic with stochastic gradients matches the sample complexity of deterministic policy gradient, reaching an $\\epsilon$-optimal regularized value with $\\tilde{O}(\\log(1/\\epsilon))$ samples. In practice, the critic is learned alongside the actor: the variance of the actor update is then influenced by the critic's variance and bias"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24357","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/2605.24357/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":"2605.24357","created_at":"2026-05-26T01:03:01.668587+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.24357v1","created_at":"2026-05-26T01:03:01.668587+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24357","created_at":"2026-05-26T01:03:01.668587+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZVXQMPBXUKAG","created_at":"2026-05-26T01:03:01.668587+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZVXQMPBXUKAGDRDI","created_at":"2026-05-26T01:03:01.668587+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZVXQMPBX","created_at":"2026-05-26T01:03:01.668587+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/ZVXQMPBXUKAGDRDINYPSZTDIBZ","json":"https://pith.science/pith/ZVXQMPBXUKAGDRDINYPSZTDIBZ.json","graph_json":"https://pith.science/api/pith-number/ZVXQMPBXUKAGDRDINYPSZTDIBZ/graph.json","events_json":"https://pith.science/api/pith-number/ZVXQMPBXUKAGDRDINYPSZTDIBZ/events.json","paper":"https://pith.science/paper/ZVXQMPBX"},"agent_actions":{"view_html":"https://pith.science/pith/ZVXQMPBXUKAGDRDINYPSZTDIBZ","download_json":"https://pith.science/pith/ZVXQMPBXUKAGDRDINYPSZTDIBZ.json","view_paper":"https://pith.science/paper/ZVXQMPBX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.24357&json=true","fetch_graph":"https://pith.science/api/pith-number/ZVXQMPBXUKAGDRDINYPSZTDIBZ/graph.json","fetch_events":"https://pith.science/api/pith-number/ZVXQMPBXUKAGDRDINYPSZTDIBZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZVXQMPBXUKAGDRDINYPSZTDIBZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZVXQMPBXUKAGDRDINYPSZTDIBZ/action/storage_attestation","attest_author":"https://pith.science/pith/ZVXQMPBXUKAGDRDINYPSZTDIBZ/action/author_attestation","sign_citation":"https://pith.science/pith/ZVXQMPBXUKAGDRDINYPSZTDIBZ/action/citation_signature","submit_replication":"https://pith.science/pith/ZVXQMPBXUKAGDRDINYPSZTDIBZ/action/replication_record"}},"created_at":"2026-05-26T01:03:01.668587+00:00","updated_at":"2026-05-26T01:03:01.668587+00:00"}