{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:JW67U2SAZQLZJ3IMQDZL7AP2TO","short_pith_number":"pith:JW67U2SA","schema_version":"1.0","canonical_sha256":"4dbdfa6a40cc1794ed0c80f2bf81fa9ba49f5fbed95d0e0eccd39680538f4fe7","source":{"kind":"arxiv","id":"2209.08429","version":1},"attestation_state":"computed","paper":{"title":"Constrained Policy Optimization for Controlled Self-Learning in Conversational AI Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Mohammad Kachuee, Sungjin Lee","submitted_at":"2022-09-17T23:44:13Z","abstract_excerpt":"Recently, self-learning methods based on user satisfaction metrics and contextual bandits have shown promising results to enable consistent improvements in conversational AI systems. However, directly targeting such metrics by off-policy bandit learning objectives often increases the risk of making abrupt policy changes that break the current user experience. In this study, we introduce a scalable framework for supporting fine-grained exploration targets for individual domains via user-defined constraints. For example, we may want to ensure fewer policy deviations in business-critical domains "},"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":"2209.08429","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-09-17T23:44:13Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"b747c4dd65589c355c34ea0d49aca9d3337a875fd41bf3ebd68d5805c0107168","abstract_canon_sha256":"eccbb1a8fe15f92185f88e6fb12dffc8339cb2062165d4a43fc52bdaa37308fb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:09:44.947895Z","signature_b64":"bt7BbgYogZ5kKsJW7r8OvYkOCCKRt0SsbIF0fu8kUtmzqjKv7ptHjBhdVUiTpIca1p46OrBHChCh14jEAo2JBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4dbdfa6a40cc1794ed0c80f2bf81fa9ba49f5fbed95d0e0eccd39680538f4fe7","last_reissued_at":"2026-07-05T06:09:44.947431Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:09:44.947431Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Constrained Policy Optimization for Controlled Self-Learning in Conversational AI Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Mohammad Kachuee, Sungjin Lee","submitted_at":"2022-09-17T23:44:13Z","abstract_excerpt":"Recently, self-learning methods based on user satisfaction metrics and contextual bandits have shown promising results to enable consistent improvements in conversational AI systems. However, directly targeting such metrics by off-policy bandit learning objectives often increases the risk of making abrupt policy changes that break the current user experience. In this study, we introduce a scalable framework for supporting fine-grained exploration targets for individual domains via user-defined constraints. For example, we may want to ensure fewer policy deviations in business-critical domains "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.08429","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/2209.08429/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":"2209.08429","created_at":"2026-07-05T06:09:44.947490+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.08429v1","created_at":"2026-07-05T06:09:44.947490+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.08429","created_at":"2026-07-05T06:09:44.947490+00:00"},{"alias_kind":"pith_short_12","alias_value":"JW67U2SAZQLZ","created_at":"2026-07-05T06:09:44.947490+00:00"},{"alias_kind":"pith_short_16","alias_value":"JW67U2SAZQLZJ3IM","created_at":"2026-07-05T06:09:44.947490+00:00"},{"alias_kind":"pith_short_8","alias_value":"JW67U2SA","created_at":"2026-07-05T06:09:44.947490+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/JW67U2SAZQLZJ3IMQDZL7AP2TO","json":"https://pith.science/pith/JW67U2SAZQLZJ3IMQDZL7AP2TO.json","graph_json":"https://pith.science/api/pith-number/JW67U2SAZQLZJ3IMQDZL7AP2TO/graph.json","events_json":"https://pith.science/api/pith-number/JW67U2SAZQLZJ3IMQDZL7AP2TO/events.json","paper":"https://pith.science/paper/JW67U2SA"},"agent_actions":{"view_html":"https://pith.science/pith/JW67U2SAZQLZJ3IMQDZL7AP2TO","download_json":"https://pith.science/pith/JW67U2SAZQLZJ3IMQDZL7AP2TO.json","view_paper":"https://pith.science/paper/JW67U2SA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.08429&json=true","fetch_graph":"https://pith.science/api/pith-number/JW67U2SAZQLZJ3IMQDZL7AP2TO/graph.json","fetch_events":"https://pith.science/api/pith-number/JW67U2SAZQLZJ3IMQDZL7AP2TO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JW67U2SAZQLZJ3IMQDZL7AP2TO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JW67U2SAZQLZJ3IMQDZL7AP2TO/action/storage_attestation","attest_author":"https://pith.science/pith/JW67U2SAZQLZJ3IMQDZL7AP2TO/action/author_attestation","sign_citation":"https://pith.science/pith/JW67U2SAZQLZJ3IMQDZL7AP2TO/action/citation_signature","submit_replication":"https://pith.science/pith/JW67U2SAZQLZJ3IMQDZL7AP2TO/action/replication_record"}},"created_at":"2026-07-05T06:09:44.947490+00:00","updated_at":"2026-07-05T06:09:44.947490+00:00"}