{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:NRHJVHY2LLGPDUAI677JUXPBVH","short_pith_number":"pith:NRHJVHY2","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"},"canonical_sha256":"6c4e9a9f1a5accf1d008f7fe9a5de1a9e7cb8f4065350b056995e05590058657","source":{"kind":"arxiv","id":"1910.05876","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1910.05876","created_at":"2026-07-05T00:11:42Z"},{"alias_kind":"arxiv_version","alias_value":"1910.05876v1","created_at":"2026-07-05T00:11:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.05876","created_at":"2026-07-05T00:11:42Z"},{"alias_kind":"pith_short_12","alias_value":"NRHJVHY2LLGP","created_at":"2026-07-05T00:11:42Z"},{"alias_kind":"pith_short_16","alias_value":"NRHJVHY2LLGPDUAI","created_at":"2026-07-05T00:11:42Z"},{"alias_kind":"pith_short_8","alias_value":"NRHJVHY2","created_at":"2026-07-05T00:11:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:NRHJVHY2LLGPDUAI677JUXPBVH","target":"record","payload":{"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"},"canonical_sha256":"6c4e9a9f1a5accf1d008f7fe9a5de1a9e7cb8f4065350b056995e05590058657","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"},"source_kind":"arxiv","source_id":"1910.05876","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":"2026-07-05T00:11:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aD6ss/BzfRPX6LrwSv5FO+VWa+Of8V1Z79wD6MhvIalCA2RTS29nTMzsIHtWTQwu7IBqzEyf/6+IOjEcJvTpBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T07:50:27.160610Z"},"content_sha256":"ca45c6b75c82f40d733da644f73cbfb072833890a57b868309f480fb25392592","schema_version":"1.0","event_id":"sha256:ca45c6b75c82f40d733da644f73cbfb072833890a57b868309f480fb25392592"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:NRHJVHY2LLGPDUAI677JUXPBVH","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T00:11:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"l9PD0Nin8kN9cgOir3g4LSXtEESgoDkNM9JP7bzs1TtFTy0x17+TzMRuM/l+g09SGVP8ZFFBSHu4TcdDf0y3Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T07:50:27.161124Z"},"content_sha256":"051d4b3c98ef2347526336cd21ed23ef19ce7d61314387bb0a572c9fc1e0a832","schema_version":"1.0","event_id":"sha256:051d4b3c98ef2347526336cd21ed23ef19ce7d61314387bb0a572c9fc1e0a832"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH/bundle.json","state_url":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NRHJVHY2LLGPDUAI677JUXPBVH/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":"2026-08-06T07:50:27Z","links":{"resolver":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH","bundle":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH/bundle.json","state":"https://pith.science/pith/NRHJVHY2LLGPDUAI677JUXPBVH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NRHJVHY2LLGPDUAI677JUXPBVH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:NRHJVHY2LLGPDUAI677JUXPBVH","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":"b06f9456a8fe31cdb172b86df3c2df921e4d5ac5e81baf851c2b4420a4fd7e4f","cross_cats_sorted":["stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2019-10-14T01:19:55Z","title_canon_sha256":"03ec78515a64479ea40b42609c01446e0e48e56f33a73a9fd62c1fe9e90c1bbe"},"schema_version":"1.0","source":{"id":"1910.05876","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1910.05876","created_at":"2026-07-05T00:11:42Z"},{"alias_kind":"arxiv_version","alias_value":"1910.05876v1","created_at":"2026-07-05T00:11:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.05876","created_at":"2026-07-05T00:11:42Z"},{"alias_kind":"pith_short_12","alias_value":"NRHJVHY2LLGP","created_at":"2026-07-05T00:11:42Z"},{"alias_kind":"pith_short_16","alias_value":"NRHJVHY2LLGPDUAI","created_at":"2026-07-05T00:11:42Z"},{"alias_kind":"pith_short_8","alias_value":"NRHJVHY2","created_at":"2026-07-05T00:11:42Z"}],"graph_snapshots":[{"event_id":"sha256:051d4b3c98ef2347526336cd21ed23ef19ce7d61314387bb0a572c9fc1e0a832","target":"graph","created_at":"2026-07-05T00:11:42Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/1910.05876/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Borja Balle, Doina Precup, Jonathan Lebensold, William Hamilton","cross_cats":["stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2019-10-14T01:19:55Z","title":"Actor Critic with Differentially Private Critic"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.05876","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:ca45c6b75c82f40d733da644f73cbfb072833890a57b868309f480fb25392592","target":"record","created_at":"2026-07-05T00:11:42Z","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":"b06f9456a8fe31cdb172b86df3c2df921e4d5ac5e81baf851c2b4420a4fd7e4f","cross_cats_sorted":["stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2019-10-14T01:19:55Z","title_canon_sha256":"03ec78515a64479ea40b42609c01446e0e48e56f33a73a9fd62c1fe9e90c1bbe"},"schema_version":"1.0","source":{"id":"1910.05876","kind":"arxiv","version":1}},"canonical_sha256":"6c4e9a9f1a5accf1d008f7fe9a5de1a9e7cb8f4065350b056995e05590058657","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6c4e9a9f1a5accf1d008f7fe9a5de1a9e7cb8f4065350b056995e05590058657","first_computed_at":"2026-07-05T00:11:42.317478Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:11:42.317478Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pFL42bh2o7S4JMtmG7gWBOFcvCp0IjTTi+tCa+TXJqMdovG1t5HhxmI8yAQ6iQ0IoW79pdKMznb07MKv4LNfCw==","signature_status":"signed_v1","signed_at":"2026-07-05T00:11:42.318047Z","signed_message":"canonical_sha256_bytes"},"source_id":"1910.05876","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ca45c6b75c82f40d733da644f73cbfb072833890a57b868309f480fb25392592","sha256:051d4b3c98ef2347526336cd21ed23ef19ce7d61314387bb0a572c9fc1e0a832"],"state_sha256":"fd6e8292deb19c0996e59eb8b9e4f0398be42329d49fae9511b3546a50d88fc9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LvJqKiYgQ6HXii+OOoT91wkESHFNhYLPdHxKm46UyVNZRik2YsoALcSmaeIWASFLDUsizueHQbuLwmHV0B5mBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T07:50:27.166514Z","bundle_sha256":"da9b1ea4258450e457f74e34d3f86338a4fa1b47f76da9f74b1c2ba36acf3e3d"}}