{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:JUZXUANQK2CHCPBQM3WEXK6WXW","short_pith_number":"pith:JUZXUANQ","canonical_record":{"source":{"id":"1706.00387","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-01T17:00:52Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"d5099d2f721e787759304db2d8511a3078ab0a1f5a8c8fa1cbe07826afa21fdc","abstract_canon_sha256":"441da829df57dc6938085fd89154a1fb895c875e97bc1846e86ca17c60bfb356"},"schema_version":"1.0"},"canonical_sha256":"4d337a01b05684713c3066ec4babd6bd940aa51100e6a8e8f6bc0139d684e471","source":{"kind":"arxiv","id":"1706.00387","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1706.00387","created_at":"2026-05-18T00:43:14Z"},{"alias_kind":"arxiv_version","alias_value":"1706.00387v1","created_at":"2026-05-18T00:43:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1706.00387","created_at":"2026-05-18T00:43:14Z"},{"alias_kind":"pith_short_12","alias_value":"JUZXUANQK2CH","created_at":"2026-05-18T12:31:24Z"},{"alias_kind":"pith_short_16","alias_value":"JUZXUANQK2CHCPBQ","created_at":"2026-05-18T12:31:24Z"},{"alias_kind":"pith_short_8","alias_value":"JUZXUANQ","created_at":"2026-05-18T12:31:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:JUZXUANQK2CHCPBQM3WEXK6WXW","target":"record","payload":{"canonical_record":{"source":{"id":"1706.00387","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-01T17:00:52Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"d5099d2f721e787759304db2d8511a3078ab0a1f5a8c8fa1cbe07826afa21fdc","abstract_canon_sha256":"441da829df57dc6938085fd89154a1fb895c875e97bc1846e86ca17c60bfb356"},"schema_version":"1.0"},"canonical_sha256":"4d337a01b05684713c3066ec4babd6bd940aa51100e6a8e8f6bc0139d684e471","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:43:14.162997Z","signature_b64":"0ShOXg1HoU9dBjaoxHpOCRldlDIawwnU7B+9fIKk7L5oMkIhCcUBV6XXTYmiAaSlfDynEI1ycta0xrXxjQwNDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4d337a01b05684713c3066ec4babd6bd940aa51100e6a8e8f6bc0139d684e471","last_reissued_at":"2026-05-18T00:43:14.162296Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:43:14.162296Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1706.00387","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":"2026-05-18T00:43:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Pw7juGAaDZIBaEENtLlyDoHOyoWO2OSNooAOqdGnrK9ebB8jeHbN8A/YLWir4qfgN2mVuJ1Af6WfX5acTx5rAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T10:34:08.347910Z"},"content_sha256":"e3106f15153d0d011dac56b837a56dac9835ab9d182b5bf1136d9c5057cb17b7","schema_version":"1.0","event_id":"sha256:e3106f15153d0d011dac56b837a56dac9835ab9d182b5bf1136d9c5057cb17b7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:JUZXUANQK2CHCPBQM3WEXK6WXW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Interpolated Policy Gradient: Merging On-Policy and Off-Policy Gradient Estimation for Deep Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Bernhard Sch\\\"olkopf, Richard E. Turner, Sergey Levine, Shixiang Gu, Timothy Lillicrap, Zoubin Ghahramani","submitted_at":"2017-06-01T17:00:52Z","abstract_excerpt":"Off-policy model-free deep reinforcement learning methods using previously collected data can improve sample efficiency over on-policy policy gradient techniques. On the other hand, on-policy algorithms are often more stable and easier to use. This paper examines, both theoretically and empirically, approaches to merging on- and off-policy updates for deep reinforcement learning. Theoretical results show that off-policy updates with a value function estimator can be interpolated with on-policy policy gradient updates whilst still satisfying performance bounds. Our analysis uses control variate"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1706.00387","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":"2026-05-18T00:43:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZXrD2MifgG8wm8fPkKHoY9rzeEK48O2P+Q+FbDCEPI3dOAScWUC/TmYOChNFhgzS8ytYc8K9z1NdhWIDqGlMAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T10:34:08.348524Z"},"content_sha256":"0fefcc554b5d446a1c01a29dbe82511434416b00c9a0f955b08392aaab0e2750","schema_version":"1.0","event_id":"sha256:0fefcc554b5d446a1c01a29dbe82511434416b00c9a0f955b08392aaab0e2750"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JUZXUANQK2CHCPBQM3WEXK6WXW/bundle.json","state_url":"https://pith.science/pith/JUZXUANQK2CHCPBQM3WEXK6WXW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JUZXUANQK2CHCPBQM3WEXK6WXW/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":"2026-05-27T10:34:08Z","links":{"resolver":"https://pith.science/pith/JUZXUANQK2CHCPBQM3WEXK6WXW","bundle":"https://pith.science/pith/JUZXUANQK2CHCPBQM3WEXK6WXW/bundle.json","state":"https://pith.science/pith/JUZXUANQK2CHCPBQM3WEXK6WXW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JUZXUANQK2CHCPBQM3WEXK6WXW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:JUZXUANQK2CHCPBQM3WEXK6WXW","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":"441da829df57dc6938085fd89154a1fb895c875e97bc1846e86ca17c60bfb356","cross_cats_sorted":["cs.AI","cs.RO"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-01T17:00:52Z","title_canon_sha256":"d5099d2f721e787759304db2d8511a3078ab0a1f5a8c8fa1cbe07826afa21fdc"},"schema_version":"1.0","source":{"id":"1706.00387","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1706.00387","created_at":"2026-05-18T00:43:14Z"},{"alias_kind":"arxiv_version","alias_value":"1706.00387v1","created_at":"2026-05-18T00:43:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1706.00387","created_at":"2026-05-18T00:43:14Z"},{"alias_kind":"pith_short_12","alias_value":"JUZXUANQK2CH","created_at":"2026-05-18T12:31:24Z"},{"alias_kind":"pith_short_16","alias_value":"JUZXUANQK2CHCPBQ","created_at":"2026-05-18T12:31:24Z"},{"alias_kind":"pith_short_8","alias_value":"JUZXUANQ","created_at":"2026-05-18T12:31:24Z"}],"graph_snapshots":[{"event_id":"sha256:0fefcc554b5d446a1c01a29dbe82511434416b00c9a0f955b08392aaab0e2750","target":"graph","created_at":"2026-05-18T00:43:14Z","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":"Off-policy model-free deep reinforcement learning methods using previously collected data can improve sample efficiency over on-policy policy gradient techniques. On the other hand, on-policy algorithms are often more stable and easier to use. This paper examines, both theoretically and empirically, approaches to merging on- and off-policy updates for deep reinforcement learning. Theoretical results show that off-policy updates with a value function estimator can be interpolated with on-policy policy gradient updates whilst still satisfying performance bounds. Our analysis uses control variate","authors_text":"Bernhard Sch\\\"olkopf, Richard E. Turner, Sergey Levine, Shixiang Gu, Timothy Lillicrap, Zoubin Ghahramani","cross_cats":["cs.AI","cs.RO"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-01T17:00:52Z","title":"Interpolated Policy Gradient: Merging On-Policy and Off-Policy Gradient Estimation for Deep Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1706.00387","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:e3106f15153d0d011dac56b837a56dac9835ab9d182b5bf1136d9c5057cb17b7","target":"record","created_at":"2026-05-18T00:43:14Z","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":"441da829df57dc6938085fd89154a1fb895c875e97bc1846e86ca17c60bfb356","cross_cats_sorted":["cs.AI","cs.RO"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-01T17:00:52Z","title_canon_sha256":"d5099d2f721e787759304db2d8511a3078ab0a1f5a8c8fa1cbe07826afa21fdc"},"schema_version":"1.0","source":{"id":"1706.00387","kind":"arxiv","version":1}},"canonical_sha256":"4d337a01b05684713c3066ec4babd6bd940aa51100e6a8e8f6bc0139d684e471","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4d337a01b05684713c3066ec4babd6bd940aa51100e6a8e8f6bc0139d684e471","first_computed_at":"2026-05-18T00:43:14.162296Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:43:14.162296Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0ShOXg1HoU9dBjaoxHpOCRldlDIawwnU7B+9fIKk7L5oMkIhCcUBV6XXTYmiAaSlfDynEI1ycta0xrXxjQwNDA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:43:14.162997Z","signed_message":"canonical_sha256_bytes"},"source_id":"1706.00387","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e3106f15153d0d011dac56b837a56dac9835ab9d182b5bf1136d9c5057cb17b7","sha256:0fefcc554b5d446a1c01a29dbe82511434416b00c9a0f955b08392aaab0e2750"],"state_sha256":"71a0490090af4e671cea73716224abb457b5e0b80fdf65983cb7c370342e0d78"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wAX+PRFMbrZU7m3KplAOtmgkBPabHudTe+ZsuzfvSYqHvkp5qtoneaNR4EI1rqmF7KWiulwUO9/nNgryGV+4Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T10:34:08.351722Z","bundle_sha256":"8738401ae016f6ec9bf93014fa0f8827925a497d07726e19580a976bacf13a90"}}