{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7TOOUNCGGLRGIKUEFNKBECBQAZ","short_pith_number":"pith:7TOOUNCG","schema_version":"1.0","canonical_sha256":"fcdcea344632e2642a842b54120830066b76b3966da8625ae597d4e6946489c2","source":{"kind":"arxiv","id":"2505.04553","version":2},"attestation_state":"computed","paper":{"title":"Risk-sensitive Reinforcement Learning Based on Convex Scoring Functions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","q-fin.RM"],"primary_cat":"q-fin.MF","authors_text":"Shanyu Han, Xiang Yu, Yang Liu","submitted_at":"2025-05-07T16:31:42Z","abstract_excerpt":"We propose a reinforcement learning (RL) framework under a broad class of risk objectives, characterized by convex scoring functions. This class covers many common risk measures, such as variance, Expected Shortfall, entropic Value-at-Risk, and mean-risk utility. To resolve the time-inconsistency issue, we consider an augmented state space and an auxiliary variable and recast the problem as a two-state optimization problem. We propose a customized Actor-Critic algorithm and establish some theoretical approximation guarantees. A key theoretical contribution is that our results do not require th"},"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":"2505.04553","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"q-fin.MF","submitted_at":"2025-05-07T16:31:42Z","cross_cats_sorted":["cs.AI","q-fin.RM"],"title_canon_sha256":"9b7362a45fac4010395a850a9aee886f21da44102020596fa73093e223dc238a","abstract_canon_sha256":"f0dd8c89c875e52ec114aab20b0d933701051f3be195332c37fd2c108f7ad608"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:03:24.158796Z","signature_b64":"DD1liQY+6sJjqMvrLwtnH9UpONIxPL3UI6ARZOkmhxEJq2JU/Vr4oyKemkyvZ8wlK4wJYgJN056X8gcSQYu7Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fcdcea344632e2642a842b54120830066b76b3966da8625ae597d4e6946489c2","last_reissued_at":"2026-07-05T11:03:24.158293Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:03:24.158293Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Risk-sensitive Reinforcement Learning Based on Convex Scoring Functions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","q-fin.RM"],"primary_cat":"q-fin.MF","authors_text":"Shanyu Han, Xiang Yu, Yang Liu","submitted_at":"2025-05-07T16:31:42Z","abstract_excerpt":"We propose a reinforcement learning (RL) framework under a broad class of risk objectives, characterized by convex scoring functions. This class covers many common risk measures, such as variance, Expected Shortfall, entropic Value-at-Risk, and mean-risk utility. To resolve the time-inconsistency issue, we consider an augmented state space and an auxiliary variable and recast the problem as a two-state optimization problem. We propose a customized Actor-Critic algorithm and establish some theoretical approximation guarantees. A key theoretical contribution is that our results do not require th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.04553","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/2505.04553/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":"2505.04553","created_at":"2026-07-05T11:03:24.158366+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.04553v2","created_at":"2026-07-05T11:03:24.158366+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.04553","created_at":"2026-07-05T11:03:24.158366+00:00"},{"alias_kind":"pith_short_12","alias_value":"7TOOUNCGGLRG","created_at":"2026-07-05T11:03:24.158366+00:00"},{"alias_kind":"pith_short_16","alias_value":"7TOOUNCGGLRGIKUE","created_at":"2026-07-05T11:03:24.158366+00:00"},{"alias_kind":"pith_short_8","alias_value":"7TOOUNCG","created_at":"2026-07-05T11:03:24.158366+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18531","citing_title":"When Does Trajectory-Level Supervision Permit Efficient Offline Reinforcement Learning?","ref_index":38,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7TOOUNCGGLRGIKUEFNKBECBQAZ","json":"https://pith.science/pith/7TOOUNCGGLRGIKUEFNKBECBQAZ.json","graph_json":"https://pith.science/api/pith-number/7TOOUNCGGLRGIKUEFNKBECBQAZ/graph.json","events_json":"https://pith.science/api/pith-number/7TOOUNCGGLRGIKUEFNKBECBQAZ/events.json","paper":"https://pith.science/paper/7TOOUNCG"},"agent_actions":{"view_html":"https://pith.science/pith/7TOOUNCGGLRGIKUEFNKBECBQAZ","download_json":"https://pith.science/pith/7TOOUNCGGLRGIKUEFNKBECBQAZ.json","view_paper":"https://pith.science/paper/7TOOUNCG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.04553&json=true","fetch_graph":"https://pith.science/api/pith-number/7TOOUNCGGLRGIKUEFNKBECBQAZ/graph.json","fetch_events":"https://pith.science/api/pith-number/7TOOUNCGGLRGIKUEFNKBECBQAZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7TOOUNCGGLRGIKUEFNKBECBQAZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7TOOUNCGGLRGIKUEFNKBECBQAZ/action/storage_attestation","attest_author":"https://pith.science/pith/7TOOUNCGGLRGIKUEFNKBECBQAZ/action/author_attestation","sign_citation":"https://pith.science/pith/7TOOUNCGGLRGIKUEFNKBECBQAZ/action/citation_signature","submit_replication":"https://pith.science/pith/7TOOUNCGGLRGIKUEFNKBECBQAZ/action/replication_record"}},"created_at":"2026-07-05T11:03:24.158366+00:00","updated_at":"2026-07-05T11:03:24.158366+00:00"}