{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:NRHJVHY2LLGPDUAI677JUXPBVH","short_pith_number":"pith:NRHJVHY2","schema_version":"1.0","canonical_sha256":"6c4e9a9f1a5accf1d008f7fe9a5de1a9e7cb8f4065350b056995e05590058657","source":{"kind":"arxiv","id":"1910.05876","version":1},"attestation_state":"computed","paper":{"title":"Actor Critic with Differentially Private Critic","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Borja Balle, Doina Precup, Jonathan Lebensold, William Hamilton","submitted_at":"2019-10-14T01:19:55Z","abstract_excerpt":"Reinforcement learning algorithms are known to be sample inefficient, and often performance on one task can be substantially improved by leveraging information (e.g., via pre-training) on other related tasks. In this work, we propose a technique to achieve such knowledge transfer in cases where agent trajectories contain sensitive or private information, such as in the healthcare domain. Our approach leverages a differentially private policy evaluation algorithm to initialize an actor-critic model and improve the effectiveness of learning in downstream tasks. We empirically show this technique"},"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":"1910.05876","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2019-10-14T01:19:55Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"03ec78515a64479ea40b42609c01446e0e48e56f33a73a9fd62c1fe9e90c1bbe","abstract_canon_sha256":"b06f9456a8fe31cdb172b86df3c2df921e4d5ac5e81baf851c2b4420a4fd7e4f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:11:42.318047Z","signature_b64":"pFL42bh2o7S4JMtmG7gWBOFcvCp0IjTTi+tCa+TXJqMdovG1t5HhxmI8yAQ6iQ0IoW79pdKMznb07MKv4LNfCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6c4e9a9f1a5accf1d008f7fe9a5de1a9e7cb8f4065350b056995e05590058657","last_reissued_at":"2026-07-05T00:11:42.317478Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:11:42.317478Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Actor Critic with Differentially Private Critic","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Borja Balle, Doina Precup, Jonathan Lebensold, William Hamilton","submitted_at":"2019-10-14T01:19:55Z","abstract_excerpt":"Reinforcement learning algorithms are known to be sample inefficient, and often performance on one task can be substantially improved by leveraging information (e.g., via pre-training) on other related tasks. In this work, we propose a technique to achieve such knowledge transfer in cases where agent trajectories contain sensitive or private information, such as in the healthcare domain. Our approach leverages a differentially private policy evaluation algorithm to initialize an actor-critic model and improve the effectiveness of learning in downstream tasks. We empirically show this technique"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.05876","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/1910.05876/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":"1910.05876","created_at":"2026-07-05T00:11:42.317547+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.05876v1","created_at":"2026-07-05T00:11:42.317547+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.05876","created_at":"2026-07-05T00:11:42.317547+00:00"},{"alias_kind":"pith_short_12","alias_value":"NRHJVHY2LLGP","created_at":"2026-07-05T00:11:42.317547+00:00"},{"alias_kind":"pith_short_16","alias_value":"NRHJVHY2LLGPDUAI","created_at":"2026-07-05T00:11:42.317547+00:00"},{"alias_kind":"pith_short_8","alias_value":"NRHJVHY2","created_at":"2026-07-05T00:11:42.317547+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/NRHJVHY2LLGPDUAI677JUXPBVH","json":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH.json","graph_json":"https://pith.science/api/pith-number/NRHJVHY2LLGPDUAI677JUXPBVH/graph.json","events_json":"https://pith.science/api/pith-number/NRHJVHY2LLGPDUAI677JUXPBVH/events.json","paper":"https://pith.science/paper/NRHJVHY2"},"agent_actions":{"view_html":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH","download_json":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH.json","view_paper":"https://pith.science/paper/NRHJVHY2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.05876&json=true","fetch_graph":"https://pith.science/api/pith-number/NRHJVHY2LLGPDUAI677JUXPBVH/graph.json","fetch_events":"https://pith.science/api/pith-number/NRHJVHY2LLGPDUAI677JUXPBVH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH/action/storage_attestation","attest_author":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH/action/author_attestation","sign_citation":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH/action/citation_signature","submit_replication":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH/action/replication_record"}},"created_at":"2026-07-05T00:11:42.317547+00:00","updated_at":"2026-07-05T00:11:42.317547+00:00"}