{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:F6YZ6QTRVICDUPMSLLXZSDP3EF","short_pith_number":"pith:F6YZ6QTR","schema_version":"1.0","canonical_sha256":"2fb19f4271aa043a3d925aef990dfb215adba88865d5b339b5c40a1a879c7beb","source":{"kind":"arxiv","id":"1707.00130","version":2},"attestation_state":"computed","paper":{"title":"Sample-efficient Actor-Critic Reinforcement Learning with Supervised Data for Dialogue Management","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Milica Gasic, Pawel Budzianowski, Pei-Hao Su, Stefan Ultes, Steve Young","submitted_at":"2017-07-01T09:56:31Z","abstract_excerpt":"Deep reinforcement learning (RL) methods have significant potential for dialogue policy optimisation. However, they suffer from a poor performance in the early stages of learning. This is especially problematic for on-line learning with real users. Two approaches are introduced to tackle this problem. Firstly, to speed up the learning process, two sample-efficient neural networks algorithms: trust region actor-critic with experience replay (TRACER) and episodic natural actor-critic with experience replay (eNACER) are presented. For TRACER, the trust region helps to control the learning step si"},"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":"1707.00130","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-07-01T09:56:31Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"8f5e795c2a414d36b488038ae38af404e8782f4b34543f585497537326a61edc","abstract_canon_sha256":"9aa329c46abb3616a385ea63e2d8a19eac9546bb6368f9265553c2e0956ee745"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:40:53.680886Z","signature_b64":"3SMuC8lhZOT9yCvp/6+lM1zl35s3M0tvuEJJf2acUJSG8Vg71v8NTKB4+t4KBSXYLn7/9vHsmlJfpEnVV2zSAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2fb19f4271aa043a3d925aef990dfb215adba88865d5b339b5c40a1a879c7beb","last_reissued_at":"2026-05-18T00:40:53.680383Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:40:53.680383Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sample-efficient Actor-Critic Reinforcement Learning with Supervised Data for Dialogue Management","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Milica Gasic, Pawel Budzianowski, Pei-Hao Su, Stefan Ultes, Steve Young","submitted_at":"2017-07-01T09:56:31Z","abstract_excerpt":"Deep reinforcement learning (RL) methods have significant potential for dialogue policy optimisation. However, they suffer from a poor performance in the early stages of learning. This is especially problematic for on-line learning with real users. Two approaches are introduced to tackle this problem. Firstly, to speed up the learning process, two sample-efficient neural networks algorithms: trust region actor-critic with experience replay (TRACER) and episodic natural actor-critic with experience replay (eNACER) are presented. For TRACER, the trust region helps to control the learning step si"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1707.00130","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":""},"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":"1707.00130","created_at":"2026-05-18T00:40:53.680468+00:00"},{"alias_kind":"arxiv_version","alias_value":"1707.00130v2","created_at":"2026-05-18T00:40:53.680468+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1707.00130","created_at":"2026-05-18T00:40:53.680468+00:00"},{"alias_kind":"pith_short_12","alias_value":"F6YZ6QTRVICD","created_at":"2026-05-18T12:31:15.632608+00:00"},{"alias_kind":"pith_short_16","alias_value":"F6YZ6QTRVICDUPMS","created_at":"2026-05-18T12:31:15.632608+00:00"},{"alias_kind":"pith_short_8","alias_value":"F6YZ6QTR","created_at":"2026-05-18T12:31:15.632608+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/F6YZ6QTRVICDUPMSLLXZSDP3EF","json":"https://pith.science/pith/F6YZ6QTRVICDUPMSLLXZSDP3EF.json","graph_json":"https://pith.science/api/pith-number/F6YZ6QTRVICDUPMSLLXZSDP3EF/graph.json","events_json":"https://pith.science/api/pith-number/F6YZ6QTRVICDUPMSLLXZSDP3EF/events.json","paper":"https://pith.science/paper/F6YZ6QTR"},"agent_actions":{"view_html":"https://pith.science/pith/F6YZ6QTRVICDUPMSLLXZSDP3EF","download_json":"https://pith.science/pith/F6YZ6QTRVICDUPMSLLXZSDP3EF.json","view_paper":"https://pith.science/paper/F6YZ6QTR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1707.00130&json=true","fetch_graph":"https://pith.science/api/pith-number/F6YZ6QTRVICDUPMSLLXZSDP3EF/graph.json","fetch_events":"https://pith.science/api/pith-number/F6YZ6QTRVICDUPMSLLXZSDP3EF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F6YZ6QTRVICDUPMSLLXZSDP3EF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F6YZ6QTRVICDUPMSLLXZSDP3EF/action/storage_attestation","attest_author":"https://pith.science/pith/F6YZ6QTRVICDUPMSLLXZSDP3EF/action/author_attestation","sign_citation":"https://pith.science/pith/F6YZ6QTRVICDUPMSLLXZSDP3EF/action/citation_signature","submit_replication":"https://pith.science/pith/F6YZ6QTRVICDUPMSLLXZSDP3EF/action/replication_record"}},"created_at":"2026-05-18T00:40:53.680468+00:00","updated_at":"2026-05-18T00:40:53.680468+00:00"}