{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:N3JSUABVGOJCE7CGQIDT66ZOC6","short_pith_number":"pith:N3JSUABV","canonical_record":{"source":{"id":"1907.12439","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-29T13:59:42Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"f92f81221d7b3913b56546bd9321073cd77ab5c9bd6d8b5a8361f63c4df9d462","abstract_canon_sha256":"1d223c2cd107afd1ba55dcd381dcd779ee6ab2f3714251b6b477ef5df1ab56eb"},"schema_version":"1.0"},"canonical_sha256":"6ed32a00353392227c4682073f7b2e17afb7a2bdf38ccb3e0e28949b1eb4cc47","source":{"kind":"arxiv","id":"1907.12439","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.12439","created_at":"2026-07-05T02:40:25Z"},{"alias_kind":"arxiv_version","alias_value":"1907.12439v5","created_at":"2026-07-05T02:40:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.12439","created_at":"2026-07-05T02:40:25Z"},{"alias_kind":"pith_short_12","alias_value":"N3JSUABVGOJC","created_at":"2026-07-05T02:40:25Z"},{"alias_kind":"pith_short_16","alias_value":"N3JSUABVGOJCE7CG","created_at":"2026-07-05T02:40:25Z"},{"alias_kind":"pith_short_8","alias_value":"N3JSUABV","created_at":"2026-07-05T02:40:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:N3JSUABVGOJCE7CGQIDT66ZOC6","target":"record","payload":{"canonical_record":{"source":{"id":"1907.12439","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-29T13:59:42Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"f92f81221d7b3913b56546bd9321073cd77ab5c9bd6d8b5a8361f63c4df9d462","abstract_canon_sha256":"1d223c2cd107afd1ba55dcd381dcd779ee6ab2f3714251b6b477ef5df1ab56eb"},"schema_version":"1.0"},"canonical_sha256":"6ed32a00353392227c4682073f7b2e17afb7a2bdf38ccb3e0e28949b1eb4cc47","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:40:25.495442Z","signature_b64":"vHIfUfbuRaxfyyweC+Jh0K0LtYmjPQphNElfis1/+pD/dOxATc/uotTNemWXTeoFsoGNchZq6zwPVV/gohiNDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ed32a00353392227c4682073f7b2e17afb7a2bdf38ccb3e0e28949b1eb4cc47","last_reissued_at":"2026-07-05T02:40:25.494941Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:40:25.494941Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1907.12439","source_version":5,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:40:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GHQSkhZEoEMenjIglJGTJ+I0RI2waW2G4P+5IFsFamcpZlbq/LnMqnxiJWpCqkxGDnojQNKc09mY5MNlu2vaAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T19:05:43.501118Z"},"content_sha256":"b00dc6a4f3eb2927e03e01362ff8f30143bf295180c8093d870432d733138950","schema_version":"1.0","event_id":"sha256:b00dc6a4f3eb2927e03e01362ff8f30143bf295180c8093d870432d733138950"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:N3JSUABVGOJCE7CGQIDT66ZOC6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Hindsight Trust Region Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"David Hsu, Hanbo Zhang, Nanning Zheng, Site Bai, Xuguang Lan","submitted_at":"2019-07-29T13:59:42Z","abstract_excerpt":"Reinforcement Learning(RL) with sparse rewards is a major challenge. We propose \\emph{Hindsight Trust Region Policy Optimization}(HTRPO), a new RL algorithm that extends the highly successful TRPO algorithm with \\emph{hindsight} to tackle the challenge of sparse rewards. Hindsight refers to the algorithm's ability to learn from information across goals, including ones not intended for the current task. HTRPO leverages two main ideas. It introduces QKL, a quadratic approximation to the KL divergence constraint on the trust region, leading to reduced variance in KL divergence estimation and impr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.12439","kind":"arxiv","version":5},"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/1907.12439/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":"2026-07-05T02:40:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aW0D+qANSn3xl2Z+Q9HrcjpsGI+Fi2ETiNEkGulaJimiReIEB8p3lhaJogkrsagJota35NXTOiuRTsjAv0WFDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T19:05:43.501609Z"},"content_sha256":"e2837be94bdeceaaf76d170d6ff10b8939e2388c37948069874cea3d77cda9e4","schema_version":"1.0","event_id":"sha256:e2837be94bdeceaaf76d170d6ff10b8939e2388c37948069874cea3d77cda9e4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/N3JSUABVGOJCE7CGQIDT66ZOC6/bundle.json","state_url":"https://pith.science/pith/N3JSUABVGOJCE7CGQIDT66ZOC6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/N3JSUABVGOJCE7CGQIDT66ZOC6/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":"2026-08-11T19:05:43Z","links":{"resolver":"https://pith.science/pith/N3JSUABVGOJCE7CGQIDT66ZOC6","bundle":"https://pith.science/pith/N3JSUABVGOJCE7CGQIDT66ZOC6/bundle.json","state":"https://pith.science/pith/N3JSUABVGOJCE7CGQIDT66ZOC6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/N3JSUABVGOJCE7CGQIDT66ZOC6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:N3JSUABVGOJCE7CGQIDT66ZOC6","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":"1d223c2cd107afd1ba55dcd381dcd779ee6ab2f3714251b6b477ef5df1ab56eb","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-29T13:59:42Z","title_canon_sha256":"f92f81221d7b3913b56546bd9321073cd77ab5c9bd6d8b5a8361f63c4df9d462"},"schema_version":"1.0","source":{"id":"1907.12439","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.12439","created_at":"2026-07-05T02:40:25Z"},{"alias_kind":"arxiv_version","alias_value":"1907.12439v5","created_at":"2026-07-05T02:40:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.12439","created_at":"2026-07-05T02:40:25Z"},{"alias_kind":"pith_short_12","alias_value":"N3JSUABVGOJC","created_at":"2026-07-05T02:40:25Z"},{"alias_kind":"pith_short_16","alias_value":"N3JSUABVGOJCE7CG","created_at":"2026-07-05T02:40:25Z"},{"alias_kind":"pith_short_8","alias_value":"N3JSUABV","created_at":"2026-07-05T02:40:25Z"}],"graph_snapshots":[{"event_id":"sha256:e2837be94bdeceaaf76d170d6ff10b8939e2388c37948069874cea3d77cda9e4","target":"graph","created_at":"2026-07-05T02:40:25Z","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/1907.12439/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement Learning(RL) with sparse rewards is a major challenge. We propose \\emph{Hindsight Trust Region Policy Optimization}(HTRPO), a new RL algorithm that extends the highly successful TRPO algorithm with \\emph{hindsight} to tackle the challenge of sparse rewards. Hindsight refers to the algorithm's ability to learn from information across goals, including ones not intended for the current task. HTRPO leverages two main ideas. It introduces QKL, a quadratic approximation to the KL divergence constraint on the trust region, leading to reduced variance in KL divergence estimation and impr","authors_text":"David Hsu, Hanbo Zhang, Nanning Zheng, Site Bai, Xuguang Lan","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-29T13:59:42Z","title":"Hindsight Trust Region Policy Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.12439","kind":"arxiv","version":5},"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:b00dc6a4f3eb2927e03e01362ff8f30143bf295180c8093d870432d733138950","target":"record","created_at":"2026-07-05T02:40:25Z","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":"1d223c2cd107afd1ba55dcd381dcd779ee6ab2f3714251b6b477ef5df1ab56eb","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-29T13:59:42Z","title_canon_sha256":"f92f81221d7b3913b56546bd9321073cd77ab5c9bd6d8b5a8361f63c4df9d462"},"schema_version":"1.0","source":{"id":"1907.12439","kind":"arxiv","version":5}},"canonical_sha256":"6ed32a00353392227c4682073f7b2e17afb7a2bdf38ccb3e0e28949b1eb4cc47","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6ed32a00353392227c4682073f7b2e17afb7a2bdf38ccb3e0e28949b1eb4cc47","first_computed_at":"2026-07-05T02:40:25.494941Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:40:25.494941Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vHIfUfbuRaxfyyweC+Jh0K0LtYmjPQphNElfis1/+pD/dOxATc/uotTNemWXTeoFsoGNchZq6zwPVV/gohiNDw==","signature_status":"signed_v1","signed_at":"2026-07-05T02:40:25.495442Z","signed_message":"canonical_sha256_bytes"},"source_id":"1907.12439","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b00dc6a4f3eb2927e03e01362ff8f30143bf295180c8093d870432d733138950","sha256:e2837be94bdeceaaf76d170d6ff10b8939e2388c37948069874cea3d77cda9e4"],"state_sha256":"8592edefa5fc0445f7de18689166873c032bfcdbda9c74a639003ee8c7006941"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0ouub3Vbx2KpCWqDt3qsLS0CBnWHedgjtkFRjRgItHsFXxXUQ8A0ZYlCAE5OzXlOmSp+wGA3ojy0rJB3ui7QBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T19:05:43.505112Z","bundle_sha256":"b8d8a5bab94bc16e3f78ad5691f416675ff91b244c64c1a02dbce1701ea8dcf3"}}