{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:J6F5237MLDIV3HWZRT4SFJWHXB","short_pith_number":"pith:J6F5237M","schema_version":"1.0","canonical_sha256":"4f8bdd6fec58d15d9ed98cf922a6c7b8636784f71e90b127c936cc60174324e3","source":{"kind":"arxiv","id":"2210.13011","version":5},"attestation_state":"computed","paper":{"title":"On Many-Actions Policy Gradient","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Marek Cygan, Michal Nauman","submitted_at":"2022-10-24T07:58:48Z","abstract_excerpt":"We study the variance of stochastic policy gradients (SPGs) with many action samples per state. We derive a many-actions optimality condition, which determines when many-actions SPG yields lower variance as compared to a single-action agent with proportionally extended trajectory. We propose Model-Based Many-Actions (MBMA), an approach leveraging dynamics models for many-actions sampling in the context of SPG. MBMA addresses issues associated with existing implementations of many-actions SPG and yields lower bias and comparable variance to SPG estimated from states in model-simulated rollouts."},"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":"2210.13011","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-10-24T07:58:48Z","cross_cats_sorted":[],"title_canon_sha256":"0ad3afcaa54f51ce72685fc4b148479e45593c85217cd2f7aa5b6705bb32ca90","abstract_canon_sha256":"484e7d62e8a3ef61bffa105f1416c51f7b543a67690b3d8c6476838396ab972d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:07:05.226131Z","signature_b64":"5LT6cJhr+zUPCLK/+V8lL/etEF48IfxF1CWBkWoVGkAvZ8HrWYRldt6SKDO7u+RpSf2V7UN6AP5BBjHJmEnSDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f8bdd6fec58d15d9ed98cf922a6c7b8636784f71e90b127c936cc60174324e3","last_reissued_at":"2026-07-05T07:07:05.225736Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:07:05.225736Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Many-Actions Policy Gradient","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Marek Cygan, Michal Nauman","submitted_at":"2022-10-24T07:58:48Z","abstract_excerpt":"We study the variance of stochastic policy gradients (SPGs) with many action samples per state. We derive a many-actions optimality condition, which determines when many-actions SPG yields lower variance as compared to a single-action agent with proportionally extended trajectory. We propose Model-Based Many-Actions (MBMA), an approach leveraging dynamics models for many-actions sampling in the context of SPG. MBMA addresses issues associated with existing implementations of many-actions SPG and yields lower bias and comparable variance to SPG estimated from states in model-simulated rollouts."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.13011","kind":"arxiv","version":5},"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/2210.13011/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":"2210.13011","created_at":"2026-07-05T07:07:05.225789+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.13011v5","created_at":"2026-07-05T07:07:05.225789+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.13011","created_at":"2026-07-05T07:07:05.225789+00:00"},{"alias_kind":"pith_short_12","alias_value":"J6F5237MLDIV","created_at":"2026-07-05T07:07:05.225789+00:00"},{"alias_kind":"pith_short_16","alias_value":"J6F5237MLDIV3HWZ","created_at":"2026-07-05T07:07:05.225789+00:00"},{"alias_kind":"pith_short_8","alias_value":"J6F5237M","created_at":"2026-07-05T07:07:05.225789+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/J6F5237MLDIV3HWZRT4SFJWHXB","json":"https://pith.science/pith/J6F5237MLDIV3HWZRT4SFJWHXB.json","graph_json":"https://pith.science/api/pith-number/J6F5237MLDIV3HWZRT4SFJWHXB/graph.json","events_json":"https://pith.science/api/pith-number/J6F5237MLDIV3HWZRT4SFJWHXB/events.json","paper":"https://pith.science/paper/J6F5237M"},"agent_actions":{"view_html":"https://pith.science/pith/J6F5237MLDIV3HWZRT4SFJWHXB","download_json":"https://pith.science/pith/J6F5237MLDIV3HWZRT4SFJWHXB.json","view_paper":"https://pith.science/paper/J6F5237M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.13011&json=true","fetch_graph":"https://pith.science/api/pith-number/J6F5237MLDIV3HWZRT4SFJWHXB/graph.json","fetch_events":"https://pith.science/api/pith-number/J6F5237MLDIV3HWZRT4SFJWHXB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J6F5237MLDIV3HWZRT4SFJWHXB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J6F5237MLDIV3HWZRT4SFJWHXB/action/storage_attestation","attest_author":"https://pith.science/pith/J6F5237MLDIV3HWZRT4SFJWHXB/action/author_attestation","sign_citation":"https://pith.science/pith/J6F5237MLDIV3HWZRT4SFJWHXB/action/citation_signature","submit_replication":"https://pith.science/pith/J6F5237MLDIV3HWZRT4SFJWHXB/action/replication_record"}},"created_at":"2026-07-05T07:07:05.225789+00:00","updated_at":"2026-07-05T07:07:05.225789+00:00"}