{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GWDVDW6YUZ36KJ2ZBPURVNSVN6","short_pith_number":"pith:GWDVDW6Y","canonical_record":{"source":{"id":"2412.11138","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-15T10:05:23Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cc5969154481554a078548077abc7906410d34a6a52dcb80809602b921a00fb7","abstract_canon_sha256":"1c3942967928cb2ab297e256f115991f7fab5062d3387ab565f5b1af776451cc"},"schema_version":"1.0"},"canonical_sha256":"358751dbd8a677e527590be91ab6556fbe68bd889fdd5e83d15eb47c280bb265","source":{"kind":"arxiv","id":"2412.11138","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.11138","created_at":"2026-07-05T09:49:28Z"},{"alias_kind":"arxiv_version","alias_value":"2412.11138v1","created_at":"2026-07-05T09:49:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.11138","created_at":"2026-07-05T09:49:28Z"},{"alias_kind":"pith_short_12","alias_value":"GWDVDW6YUZ36","created_at":"2026-07-05T09:49:28Z"},{"alias_kind":"pith_short_16","alias_value":"GWDVDW6YUZ36KJ2Z","created_at":"2026-07-05T09:49:28Z"},{"alias_kind":"pith_short_8","alias_value":"GWDVDW6Y","created_at":"2026-07-05T09:49:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GWDVDW6YUZ36KJ2ZBPURVNSVN6","target":"record","payload":{"canonical_record":{"source":{"id":"2412.11138","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-15T10:05:23Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cc5969154481554a078548077abc7906410d34a6a52dcb80809602b921a00fb7","abstract_canon_sha256":"1c3942967928cb2ab297e256f115991f7fab5062d3387ab565f5b1af776451cc"},"schema_version":"1.0"},"canonical_sha256":"358751dbd8a677e527590be91ab6556fbe68bd889fdd5e83d15eb47c280bb265","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:49:28.027548Z","signature_b64":"eKIfAbej0HlSxkjUufpG8y0ko9ACPquEe6ZJlNnUohdAjcb7ebPXCvY0LFDqnKIfRrtmmQ2oB3qbErlFg+QHCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"358751dbd8a677e527590be91ab6556fbe68bd889fdd5e83d15eb47c280bb265","last_reissued_at":"2026-07-05T09:49:28.027158Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:49:28.027158Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.11138","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":"2026-07-05T09:49:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mG4cv4ZBy9VuItpJG5DJ8Fgqw1KE853N6ujpanRIr3/9bin7q93/cdN2+Vi1fWiDa8lnTBvqcZndw6/bOGaVDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T22:23:19.022311Z"},"content_sha256":"aba9f52eccf7f901e91483b545189988263f1edd00549a57dd27aea38c5fd8fd","schema_version":"1.0","event_id":"sha256:aba9f52eccf7f901e91483b545189988263f1edd00549a57dd27aea38c5fd8fd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GWDVDW6YUZ36KJ2ZBPURVNSVN6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Safe Reinforcement Learning using Finite-Horizon Gradient-based Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Gang Pan, Juntao Dai, Qian Zheng, Yaodong Yang","submitted_at":"2024-12-15T10:05:23Z","abstract_excerpt":"A key aspect of Safe Reinforcement Learning (Safe RL) involves estimating the constraint condition for the next policy, which is crucial for guiding the optimization of safe policy updates. However, the existing Advantage-based Estimation (ABE) method relies on the infinite-horizon discounted advantage function. This dependence leads to catastrophic errors in finite-horizon scenarios with non-discounted constraints, resulting in safety-violation updates. In response, we propose the first estimation method for finite-horizon non-discounted constraints in deep Safe RL, termed Gradient-based Esti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.11138","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/2412.11138/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":"2026-07-05T09:49:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F78QOX2xbl+KZHzGkwqVQB8usduEXliuQy2BkkrtAKR+DpmSnLfAwOLX7uUH7BOu9A3C//4GgSmAxnVYEZe3Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T22:23:19.022836Z"},"content_sha256":"3194daebb06d19b08e1af929b365143cf6b9f2dc601a7988ccce6eb5f2a1c351","schema_version":"1.0","event_id":"sha256:3194daebb06d19b08e1af929b365143cf6b9f2dc601a7988ccce6eb5f2a1c351"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GWDVDW6YUZ36KJ2ZBPURVNSVN6/bundle.json","state_url":"https://pith.science/pith/GWDVDW6YUZ36KJ2ZBPURVNSVN6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GWDVDW6YUZ36KJ2ZBPURVNSVN6/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":"2026-08-14T22:23:19Z","links":{"resolver":"https://pith.science/pith/GWDVDW6YUZ36KJ2ZBPURVNSVN6","bundle":"https://pith.science/pith/GWDVDW6YUZ36KJ2ZBPURVNSVN6/bundle.json","state":"https://pith.science/pith/GWDVDW6YUZ36KJ2ZBPURVNSVN6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GWDVDW6YUZ36KJ2ZBPURVNSVN6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GWDVDW6YUZ36KJ2ZBPURVNSVN6","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":"1c3942967928cb2ab297e256f115991f7fab5062d3387ab565f5b1af776451cc","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-15T10:05:23Z","title_canon_sha256":"cc5969154481554a078548077abc7906410d34a6a52dcb80809602b921a00fb7"},"schema_version":"1.0","source":{"id":"2412.11138","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.11138","created_at":"2026-07-05T09:49:28Z"},{"alias_kind":"arxiv_version","alias_value":"2412.11138v1","created_at":"2026-07-05T09:49:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.11138","created_at":"2026-07-05T09:49:28Z"},{"alias_kind":"pith_short_12","alias_value":"GWDVDW6YUZ36","created_at":"2026-07-05T09:49:28Z"},{"alias_kind":"pith_short_16","alias_value":"GWDVDW6YUZ36KJ2Z","created_at":"2026-07-05T09:49:28Z"},{"alias_kind":"pith_short_8","alias_value":"GWDVDW6Y","created_at":"2026-07-05T09:49:28Z"}],"graph_snapshots":[{"event_id":"sha256:3194daebb06d19b08e1af929b365143cf6b9f2dc601a7988ccce6eb5f2a1c351","target":"graph","created_at":"2026-07-05T09:49:28Z","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/2412.11138/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A key aspect of Safe Reinforcement Learning (Safe RL) involves estimating the constraint condition for the next policy, which is crucial for guiding the optimization of safe policy updates. However, the existing Advantage-based Estimation (ABE) method relies on the infinite-horizon discounted advantage function. This dependence leads to catastrophic errors in finite-horizon scenarios with non-discounted constraints, resulting in safety-violation updates. In response, we propose the first estimation method for finite-horizon non-discounted constraints in deep Safe RL, termed Gradient-based Esti","authors_text":"Gang Pan, Juntao Dai, Qian Zheng, Yaodong Yang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-15T10:05:23Z","title":"Safe Reinforcement Learning using Finite-Horizon Gradient-based Estimation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.11138","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:aba9f52eccf7f901e91483b545189988263f1edd00549a57dd27aea38c5fd8fd","target":"record","created_at":"2026-07-05T09:49:28Z","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":"1c3942967928cb2ab297e256f115991f7fab5062d3387ab565f5b1af776451cc","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-15T10:05:23Z","title_canon_sha256":"cc5969154481554a078548077abc7906410d34a6a52dcb80809602b921a00fb7"},"schema_version":"1.0","source":{"id":"2412.11138","kind":"arxiv","version":1}},"canonical_sha256":"358751dbd8a677e527590be91ab6556fbe68bd889fdd5e83d15eb47c280bb265","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"358751dbd8a677e527590be91ab6556fbe68bd889fdd5e83d15eb47c280bb265","first_computed_at":"2026-07-05T09:49:28.027158Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:49:28.027158Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eKIfAbej0HlSxkjUufpG8y0ko9ACPquEe6ZJlNnUohdAjcb7ebPXCvY0LFDqnKIfRrtmmQ2oB3qbErlFg+QHCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:49:28.027548Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.11138","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:aba9f52eccf7f901e91483b545189988263f1edd00549a57dd27aea38c5fd8fd","sha256:3194daebb06d19b08e1af929b365143cf6b9f2dc601a7988ccce6eb5f2a1c351"],"state_sha256":"ebe7532cfafcaf047878b8af9d5a2eeb09e1cb836fc45f9ab3836d1cfb57f7f2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uvlrEm5R5N1WMmgttaK3q8mklPPoP5HrJBhoHmJLsaNfgOuTIICnqK0NoyIvZBQ4+7XC27uNfxozqeG0YO72CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T22:23:19.026239Z","bundle_sha256":"58673caf743cca6695bd2c284035b18721627573af821c01962eba4ad358189e"}}