{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:JSV5FVNR6QVXO46NI5L2CI3HI5","short_pith_number":"pith:JSV5FVNR","schema_version":"1.0","canonical_sha256":"4cabd2d5b1f42b7773cd4757a1236747487d2db0bb7ade2669fe29895dcdad3b","source":{"kind":"arxiv","id":"2007.04203","version":1},"attestation_state":"computed","paper":{"title":"A Natural Actor-Critic Algorithm with Downside Risk Constraints","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","q-fin.CP","q-fin.PM","stat.ML"],"primary_cat":"cs.LG","authors_text":"Rahul Savani, Thomas Spooner","submitted_at":"2020-07-08T15:44:33Z","abstract_excerpt":"Existing work on risk-sensitive reinforcement learning - both for symmetric and downside risk measures - has typically used direct Monte-Carlo estimation of policy gradients. While this approach yields unbiased gradient estimates, it also suffers from high variance and decreased sample efficiency compared to temporal-difference methods. In this paper, we study prediction and control with aversion to downside risk which we gauge by the lower partial moment of the return. We introduce a new Bellman equation that upper bounds the lower partial moment, circumventing its non-linearity. We prove tha"},"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":"2007.04203","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-07-08T15:44:33Z","cross_cats_sorted":["cs.AI","q-fin.CP","q-fin.PM","stat.ML"],"title_canon_sha256":"9b68cbfbcf067e104c936d4ca86e7244372227f016d204a91900dd65d6c780da","abstract_canon_sha256":"a290a188f5f1fc50b69d900f71a4bea75655e8be1e57fd4dd7ea89b8e7e929d5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:17:21.940538Z","signature_b64":"GWtX6oYZ7mKd3Lw81uuJ3U20qc+UnIo4FbAXhIBBwFPB899HuLuBsTajMljwH8piQL4dzDTWFGQwcWKBT+8GDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4cabd2d5b1f42b7773cd4757a1236747487d2db0bb7ade2669fe29895dcdad3b","last_reissued_at":"2026-07-05T01:17:21.939993Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:17:21.939993Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Natural Actor-Critic Algorithm with Downside Risk Constraints","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","q-fin.CP","q-fin.PM","stat.ML"],"primary_cat":"cs.LG","authors_text":"Rahul Savani, Thomas Spooner","submitted_at":"2020-07-08T15:44:33Z","abstract_excerpt":"Existing work on risk-sensitive reinforcement learning - both for symmetric and downside risk measures - has typically used direct Monte-Carlo estimation of policy gradients. While this approach yields unbiased gradient estimates, it also suffers from high variance and decreased sample efficiency compared to temporal-difference methods. In this paper, we study prediction and control with aversion to downside risk which we gauge by the lower partial moment of the return. We introduce a new Bellman equation that upper bounds the lower partial moment, circumventing its non-linearity. We prove tha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.04203","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/2007.04203/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":"2007.04203","created_at":"2026-07-05T01:17:21.940053+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.04203v1","created_at":"2026-07-05T01:17:21.940053+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.04203","created_at":"2026-07-05T01:17:21.940053+00:00"},{"alias_kind":"pith_short_12","alias_value":"JSV5FVNR6QVX","created_at":"2026-07-05T01:17:21.940053+00:00"},{"alias_kind":"pith_short_16","alias_value":"JSV5FVNR6QVXO46N","created_at":"2026-07-05T01:17:21.940053+00:00"},{"alias_kind":"pith_short_8","alias_value":"JSV5FVNR","created_at":"2026-07-05T01:17:21.940053+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/JSV5FVNR6QVXO46NI5L2CI3HI5","json":"https://pith.science/pith/JSV5FVNR6QVXO46NI5L2CI3HI5.json","graph_json":"https://pith.science/api/pith-number/JSV5FVNR6QVXO46NI5L2CI3HI5/graph.json","events_json":"https://pith.science/api/pith-number/JSV5FVNR6QVXO46NI5L2CI3HI5/events.json","paper":"https://pith.science/paper/JSV5FVNR"},"agent_actions":{"view_html":"https://pith.science/pith/JSV5FVNR6QVXO46NI5L2CI3HI5","download_json":"https://pith.science/pith/JSV5FVNR6QVXO46NI5L2CI3HI5.json","view_paper":"https://pith.science/paper/JSV5FVNR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.04203&json=true","fetch_graph":"https://pith.science/api/pith-number/JSV5FVNR6QVXO46NI5L2CI3HI5/graph.json","fetch_events":"https://pith.science/api/pith-number/JSV5FVNR6QVXO46NI5L2CI3HI5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JSV5FVNR6QVXO46NI5L2CI3HI5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JSV5FVNR6QVXO46NI5L2CI3HI5/action/storage_attestation","attest_author":"https://pith.science/pith/JSV5FVNR6QVXO46NI5L2CI3HI5/action/author_attestation","sign_citation":"https://pith.science/pith/JSV5FVNR6QVXO46NI5L2CI3HI5/action/citation_signature","submit_replication":"https://pith.science/pith/JSV5FVNR6QVXO46NI5L2CI3HI5/action/replication_record"}},"created_at":"2026-07-05T01:17:21.940053+00:00","updated_at":"2026-07-05T01:17:21.940053+00:00"}