{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6BYGOUEWWAEEI27ERN5KGXXNZH","short_pith_number":"pith:6BYGOUEW","schema_version":"1.0","canonical_sha256":"f070675096b008446be48b7aa35eedc9ed95649ca87581bfaa764e359664e183","source":{"kind":"arxiv","id":"2408.07245","version":3},"attestation_state":"computed","paper":{"title":"q-exponential family for policy optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Han Wang, Haseeb Shah, Lingwei Zhu, Martha White, Yukie Nagai","submitted_at":"2024-08-14T01:01:35Z","abstract_excerpt":"Policy optimization methods benefit from a simple and tractable policy parametrization, usually the Gaussian for continuous action spaces. In this paper, we consider a broader policy family that remains tractable: the $q$-exponential family. This family of policies is flexible, allowing the specification of both heavy-tailed policies ($q>1$) and light-tailed policies ($q<1$). This paper examines the interplay between $q$-exponential policies for several actor-critic algorithms conducted on both online and offline problems. We find that heavy-tailed policies are more effective in general and ca"},"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":"2408.07245","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-08-14T01:01:35Z","cross_cats_sorted":[],"title_canon_sha256":"f097da787f37a785bbe6275d2ded7dfd701e0f0cf40681e97038fcf9862b520e","abstract_canon_sha256":"f7f4f0e039b735e20842c9db93dc63558be96c9e4ead333c5c2c2140c256d531"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:04:57.667846Z","signature_b64":"mnOWmUjm/EBYh3c/RSm6IGA//uBaxmqVW+9CT/8PuLGzDawrqY1xjdng7g2f8c8dgxYYOjNEvV0HVf9RJZl5Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f070675096b008446be48b7aa35eedc9ed95649ca87581bfaa764e359664e183","last_reissued_at":"2026-07-05T10:04:57.667369Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:04:57.667369Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"q-exponential family for policy optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Han Wang, Haseeb Shah, Lingwei Zhu, Martha White, Yukie Nagai","submitted_at":"2024-08-14T01:01:35Z","abstract_excerpt":"Policy optimization methods benefit from a simple and tractable policy parametrization, usually the Gaussian for continuous action spaces. In this paper, we consider a broader policy family that remains tractable: the $q$-exponential family. This family of policies is flexible, allowing the specification of both heavy-tailed policies ($q>1$) and light-tailed policies ($q<1$). This paper examines the interplay between $q$-exponential policies for several actor-critic algorithms conducted on both online and offline problems. We find that heavy-tailed policies are more effective in general and ca"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.07245","kind":"arxiv","version":3},"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/2408.07245/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":"2408.07245","created_at":"2026-07-05T10:04:57.667426+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.07245v3","created_at":"2026-07-05T10:04:57.667426+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.07245","created_at":"2026-07-05T10:04:57.667426+00:00"},{"alias_kind":"pith_short_12","alias_value":"6BYGOUEWWAEE","created_at":"2026-07-05T10:04:57.667426+00:00"},{"alias_kind":"pith_short_16","alias_value":"6BYGOUEWWAEEI27E","created_at":"2026-07-05T10:04:57.667426+00:00"},{"alias_kind":"pith_short_8","alias_value":"6BYGOUEW","created_at":"2026-07-05T10:04:57.667426+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/6BYGOUEWWAEEI27ERN5KGXXNZH","json":"https://pith.science/pith/6BYGOUEWWAEEI27ERN5KGXXNZH.json","graph_json":"https://pith.science/api/pith-number/6BYGOUEWWAEEI27ERN5KGXXNZH/graph.json","events_json":"https://pith.science/api/pith-number/6BYGOUEWWAEEI27ERN5KGXXNZH/events.json","paper":"https://pith.science/paper/6BYGOUEW"},"agent_actions":{"view_html":"https://pith.science/pith/6BYGOUEWWAEEI27ERN5KGXXNZH","download_json":"https://pith.science/pith/6BYGOUEWWAEEI27ERN5KGXXNZH.json","view_paper":"https://pith.science/paper/6BYGOUEW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.07245&json=true","fetch_graph":"https://pith.science/api/pith-number/6BYGOUEWWAEEI27ERN5KGXXNZH/graph.json","fetch_events":"https://pith.science/api/pith-number/6BYGOUEWWAEEI27ERN5KGXXNZH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6BYGOUEWWAEEI27ERN5KGXXNZH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6BYGOUEWWAEEI27ERN5KGXXNZH/action/storage_attestation","attest_author":"https://pith.science/pith/6BYGOUEWWAEEI27ERN5KGXXNZH/action/author_attestation","sign_citation":"https://pith.science/pith/6BYGOUEWWAEEI27ERN5KGXXNZH/action/citation_signature","submit_replication":"https://pith.science/pith/6BYGOUEWWAEEI27ERN5KGXXNZH/action/replication_record"}},"created_at":"2026-07-05T10:04:57.667426+00:00","updated_at":"2026-07-05T10:04:57.667426+00:00"}