{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2016:L6QY7UIT6SKKW546XSZR44SIO7","short_pith_number":"pith:L6QY7UIT","canonical_record":{"source":{"id":"1612.06000","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2016-12-18T21:51:10Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"cc7804a6865daf6e81794e71e1f969320547d861ba11e6410672da60abb8a00d","abstract_canon_sha256":"6701622c700b56203c7ec133f4439aebe286a333f097b1cac64927481205a83d"},"schema_version":"1.0"},"canonical_sha256":"5fa18fd113f494ab779ebcb31e724877cfdf34a36ce2964dd2eafd83609828b9","source":{"kind":"arxiv","id":"1612.06000","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1612.06000","created_at":"2026-05-18T00:54:46Z"},{"alias_kind":"arxiv_version","alias_value":"1612.06000v1","created_at":"2026-05-18T00:54:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1612.06000","created_at":"2026-05-18T00:54:46Z"},{"alias_kind":"pith_short_12","alias_value":"L6QY7UIT6SKK","created_at":"2026-05-18T12:30:29Z"},{"alias_kind":"pith_short_16","alias_value":"L6QY7UIT6SKKW546","created_at":"2026-05-18T12:30:29Z"},{"alias_kind":"pith_short_8","alias_value":"L6QY7UIT","created_at":"2026-05-18T12:30:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2016:L6QY7UIT6SKKW546XSZR44SIO7","target":"record","payload":{"canonical_record":{"source":{"id":"1612.06000","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2016-12-18T21:51:10Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"cc7804a6865daf6e81794e71e1f969320547d861ba11e6410672da60abb8a00d","abstract_canon_sha256":"6701622c700b56203c7ec133f4439aebe286a333f097b1cac64927481205a83d"},"schema_version":"1.0"},"canonical_sha256":"5fa18fd113f494ab779ebcb31e724877cfdf34a36ce2964dd2eafd83609828b9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:54:46.156769Z","signature_b64":"VTb61ot+OLpnGJhThO3XTaLLouXkE4y9QSEgH+snfqW57UXWkADuHVO832R0InDF6IAT+xiN6opwUVLEQKzYDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5fa18fd113f494ab779ebcb31e724877cfdf34a36ce2964dd2eafd83609828b9","last_reissued_at":"2026-05-18T00:54:46.156188Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:54:46.156188Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1612.06000","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:54:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FYzg1dx5omuHhsM6eLGbCXrQamjzr2of82fNvZNvkm83qYdTI79x/9Ddbglq20NrxWOKKEuByetRxukL0w/SAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T00:46:31.567670Z"},"content_sha256":"47de11202d422eea6a5ffe1795e568fcef47e985c7568b798443ad740440757c","schema_version":"1.0","event_id":"sha256:47de11202d422eea6a5ffe1795e568fcef47e985c7568b798443ad740440757c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2016:L6QY7UIT6SKKW546XSZR44SIO7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Sample-efficient Deep Reinforcement Learning for Dialog Control","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.AI","authors_text":"Jason D. Williams, Kavosh Asadi","submitted_at":"2016-12-18T21:51:10Z","abstract_excerpt":"Representing a dialog policy as a recurrent neural network (RNN) is attractive because it handles partial observability, infers a latent representation of state, and can be optimized with supervised learning (SL) or reinforcement learning (RL). For RL, a policy gradient approach is natural, but is sample inefficient. In this paper, we present 3 methods for reducing the number of dialogs required to optimize an RNN-based dialog policy with RL. The key idea is to maintain a second RNN which predicts the value of the current policy, and to apply experience replay to both networks. On two tasks, t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1612.06000","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":""},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:54:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KNDsh/5V0gO5ANZQG2SKyW5svWs4n4zbfrUlAEd9pzskQGQ4l1V2LSRl2jKkIqe5CBR+uxr5zSV/JnfGQ3VYBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T00:46:31.568547Z"},"content_sha256":"6390bcc428cc9a2a3195b7249f1343086cf51ebc70709261bb4c8a9fff31b332","schema_version":"1.0","event_id":"sha256:6390bcc428cc9a2a3195b7249f1343086cf51ebc70709261bb4c8a9fff31b332"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/L6QY7UIT6SKKW546XSZR44SIO7/bundle.json","state_url":"https://pith.science/pith/L6QY7UIT6SKKW546XSZR44SIO7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/L6QY7UIT6SKKW546XSZR44SIO7/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-11T00:46:31Z","links":{"resolver":"https://pith.science/pith/L6QY7UIT6SKKW546XSZR44SIO7","bundle":"https://pith.science/pith/L6QY7UIT6SKKW546XSZR44SIO7/bundle.json","state":"https://pith.science/pith/L6QY7UIT6SKKW546XSZR44SIO7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/L6QY7UIT6SKKW546XSZR44SIO7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2016:L6QY7UIT6SKKW546XSZR44SIO7","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"6701622c700b56203c7ec133f4439aebe286a333f097b1cac64927481205a83d","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2016-12-18T21:51:10Z","title_canon_sha256":"cc7804a6865daf6e81794e71e1f969320547d861ba11e6410672da60abb8a00d"},"schema_version":"1.0","source":{"id":"1612.06000","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1612.06000","created_at":"2026-05-18T00:54:46Z"},{"alias_kind":"arxiv_version","alias_value":"1612.06000v1","created_at":"2026-05-18T00:54:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1612.06000","created_at":"2026-05-18T00:54:46Z"},{"alias_kind":"pith_short_12","alias_value":"L6QY7UIT6SKK","created_at":"2026-05-18T12:30:29Z"},{"alias_kind":"pith_short_16","alias_value":"L6QY7UIT6SKKW546","created_at":"2026-05-18T12:30:29Z"},{"alias_kind":"pith_short_8","alias_value":"L6QY7UIT","created_at":"2026-05-18T12:30:29Z"}],"graph_snapshots":[{"event_id":"sha256:6390bcc428cc9a2a3195b7249f1343086cf51ebc70709261bb4c8a9fff31b332","target":"graph","created_at":"2026-05-18T00:54:46Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Representing a dialog policy as a recurrent neural network (RNN) is attractive because it handles partial observability, infers a latent representation of state, and can be optimized with supervised learning (SL) or reinforcement learning (RL). For RL, a policy gradient approach is natural, but is sample inefficient. In this paper, we present 3 methods for reducing the number of dialogs required to optimize an RNN-based dialog policy with RL. The key idea is to maintain a second RNN which predicts the value of the current policy, and to apply experience replay to both networks. On two tasks, t","authors_text":"Jason D. Williams, Kavosh Asadi","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2016-12-18T21:51:10Z","title":"Sample-efficient Deep Reinforcement Learning for Dialog Control"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1612.06000","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:47de11202d422eea6a5ffe1795e568fcef47e985c7568b798443ad740440757c","target":"record","created_at":"2026-05-18T00:54:46Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"6701622c700b56203c7ec133f4439aebe286a333f097b1cac64927481205a83d","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2016-12-18T21:51:10Z","title_canon_sha256":"cc7804a6865daf6e81794e71e1f969320547d861ba11e6410672da60abb8a00d"},"schema_version":"1.0","source":{"id":"1612.06000","kind":"arxiv","version":1}},"canonical_sha256":"5fa18fd113f494ab779ebcb31e724877cfdf34a36ce2964dd2eafd83609828b9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5fa18fd113f494ab779ebcb31e724877cfdf34a36ce2964dd2eafd83609828b9","first_computed_at":"2026-05-18T00:54:46.156188Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:54:46.156188Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VTb61ot+OLpnGJhThO3XTaLLouXkE4y9QSEgH+snfqW57UXWkADuHVO832R0InDF6IAT+xiN6opwUVLEQKzYDg==","signature_status":"signed_v1","signed_at":"2026-05-18T00:54:46.156769Z","signed_message":"canonical_sha256_bytes"},"source_id":"1612.06000","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:47de11202d422eea6a5ffe1795e568fcef47e985c7568b798443ad740440757c","sha256:6390bcc428cc9a2a3195b7249f1343086cf51ebc70709261bb4c8a9fff31b332"],"state_sha256":"b274ad3e02ecd7b9352dd940cbaf6d4f5847d4eca866e295eb77503fe90a022d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NKlGOcs96nalAJjwqN+DxaLEDFmh0X1PttfQw6rQksy93MXrSQylhPks26l1Rj1q25SEC3r+uLSA4KwXn63zAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T00:46:31.573217Z","bundle_sha256":"6ad57b3ba1f40befd074af27ed3e9dffebb0ef005beaad5ddc9ed5ecf590aeaa"}}