{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:AKT4PA27VFE5BJD7YPCZ75ZTEF","short_pith_number":"pith:AKT4PA27","canonical_record":{"source":{"id":"2605.11975","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T11:31:36Z","cross_cats_sorted":[],"title_canon_sha256":"66d720cff133972aa7823b5ebaac80077487d3c568a5519b8ddfdb598a6eb03a","abstract_canon_sha256":"0a8a6fcc79b1dafa5a3fd8818da64a1986023d846c8114d99a5698cd87cbac34"},"schema_version":"1.0"},"canonical_sha256":"02a7c7835fa949d0a47fc3c59ff73321548bc7f5ee6ec7f71478e8fe76a26526","source":{"kind":"arxiv","id":"2605.11975","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.11975","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"arxiv_version","alias_value":"2605.11975v2","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11975","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_12","alias_value":"AKT4PA27VFE5","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_16","alias_value":"AKT4PA27VFE5BJD7","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_8","alias_value":"AKT4PA27","created_at":"2026-05-20T00:04:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:AKT4PA27VFE5BJD7YPCZ75ZTEF","target":"record","payload":{"canonical_record":{"source":{"id":"2605.11975","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T11:31:36Z","cross_cats_sorted":[],"title_canon_sha256":"66d720cff133972aa7823b5ebaac80077487d3c568a5519b8ddfdb598a6eb03a","abstract_canon_sha256":"0a8a6fcc79b1dafa5a3fd8818da64a1986023d846c8114d99a5698cd87cbac34"},"schema_version":"1.0"},"canonical_sha256":"02a7c7835fa949d0a47fc3c59ff73321548bc7f5ee6ec7f71478e8fe76a26526","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:36.371184Z","signature_b64":"WRqpCkg+0T2fhlfdtDikkWkvvI6Ioz9Sb/Yx5sMSylWGTPipgLkkFamoAHVzCaWltmYCkoT+ps+OsoNUWppXDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"02a7c7835fa949d0a47fc3c59ff73321548bc7f5ee6ec7f71478e8fe76a26526","last_reissued_at":"2026-05-20T00:04:36.370095Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:36.370095Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.11975","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"i4K2TzHxo+H2M0EgvnFo9viSc53Dai6cnJusZAkJg/kGuG7PiDYeArIYpaIoO04MZ5G/jZ61fnYIpbL1LmXACA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-12T07:10:49.792875Z"},"content_sha256":"275536e5930903903f96502298d493ad402b32f7177532778577fdbbed5002ce","schema_version":"1.0","event_id":"sha256:275536e5930903903f96502298d493ad402b32f7177532778577fdbbed5002ce"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:AKT4PA27VFE5BJD7YPCZ75ZTEF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Stochastic Minimum-Cost Reach-Avoid Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Reach-avoid probability certificates turn stochastic safety constraints into a surrogate objective that reinforcement learners can optimize for minimum cost.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bai Xue, Jingduo Pan, Taoran Wu, Yiling Xue","submitted_at":"2026-05-12T11:31:36Z","abstract_excerpt":"We study stochastic minimum-cost reach-avoid reinforcement learning, where an agent must satisfy a reach-avoid specification with probability at least $p$ while minimizing expected cumulative costs in stochastic environments. Existing safe and constrained reinforcement learning methods typically fail to jointly enforce probabilistic reach-avoid constraints and optimize cost in the learning setting in stochastic environments. To address this challenge, we introduce reach-avoid probability certificates (RAPCs), which identify states from which stochastic reach-avoid constraints are satisfiable. "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We establish almost sure convergence of the proposed algorithms to locally optimal policies with respect to the resulting objective.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That reach-avoid probability certificates can be computed or approximated accurately enough during learning to serve as a reliable surrogate for the true probabilistic constraint in stochastic environments.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Introduces RAPCs and a contraction Bellman operator that jointly enforce probabilistic reach-avoid constraints while minimizing expected costs in stochastic RL, with almost-sure convergence to local optima.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Reach-avoid probability certificates turn stochastic safety constraints into a surrogate objective that reinforcement learners can optimize for minimum cost.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"cb98d4fcf3484c15be6e8aa3859b7521c610e76c5361292e128e68ddb09613eb"},"source":{"id":"2605.11975","kind":"arxiv","version":2},"verdict":{"id":"dbc54e43-26df-498b-8077-da1985984ccf","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T07:23:59.690351Z","strongest_claim":"We establish almost sure convergence of the proposed algorithms to locally optimal policies with respect to the resulting objective.","one_line_summary":"Introduces RAPCs and a contraction Bellman operator that jointly enforce probabilistic reach-avoid constraints while minimizing expected costs in stochastic RL, with almost-sure convergence to local optima.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That reach-avoid probability certificates can be computed or approximated accurately enough during learning to serve as a reliable surrogate for the true probabilistic constraint in stochastic environments.","pith_extraction_headline":"Reach-avoid probability certificates turn stochastic safety constraints into a surrogate objective that reinforcement learners can optimize for minimum cost."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.11975/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T11:34:51.555285Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T09:01:17.174166Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T08:02:38.043163Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"64248693b94f85b78bf902b8a5cb22190e5b51432067cb8d1578a25192c7eb03"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"b89d607eea2972b64737389308f812ef68667cb70e74931bfba336853027a791"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"dbc54e43-26df-498b-8077-da1985984ccf"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"K7O+Ji9BnFAhBtyuW3pZXKn05ZoIujK0UgWCwpY9ZMy7CswEh05aqpgFOwC/yHlt9FJ5XsVVNYLlYg7BT//FBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-12T07:10:49.793374Z"},"content_sha256":"d25b7c4af974a6e6bd366395c846292f69eee536890a3eecca3ed6a7793635eb","schema_version":"1.0","event_id":"sha256:d25b7c4af974a6e6bd366395c846292f69eee536890a3eecca3ed6a7793635eb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AKT4PA27VFE5BJD7YPCZ75ZTEF/bundle.json","state_url":"https://pith.science/pith/AKT4PA27VFE5BJD7YPCZ75ZTEF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AKT4PA27VFE5BJD7YPCZ75ZTEF/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":"2026-06-12T07:10:49Z","links":{"resolver":"https://pith.science/pith/AKT4PA27VFE5BJD7YPCZ75ZTEF","bundle":"https://pith.science/pith/AKT4PA27VFE5BJD7YPCZ75ZTEF/bundle.json","state":"https://pith.science/pith/AKT4PA27VFE5BJD7YPCZ75ZTEF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AKT4PA27VFE5BJD7YPCZ75ZTEF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:AKT4PA27VFE5BJD7YPCZ75ZTEF","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":"0a8a6fcc79b1dafa5a3fd8818da64a1986023d846c8114d99a5698cd87cbac34","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T11:31:36Z","title_canon_sha256":"66d720cff133972aa7823b5ebaac80077487d3c568a5519b8ddfdb598a6eb03a"},"schema_version":"1.0","source":{"id":"2605.11975","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.11975","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"arxiv_version","alias_value":"2605.11975v2","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11975","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_12","alias_value":"AKT4PA27VFE5","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_16","alias_value":"AKT4PA27VFE5BJD7","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_8","alias_value":"AKT4PA27","created_at":"2026-05-20T00:04:36Z"}],"graph_snapshots":[{"event_id":"sha256:d25b7c4af974a6e6bd366395c846292f69eee536890a3eecca3ed6a7793635eb","target":"graph","created_at":"2026-05-20T00:04:36Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"We establish almost sure convergence of the proposed algorithms to locally optimal policies with respect to the resulting objective."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That reach-avoid probability certificates can be computed or approximated accurately enough during learning to serve as a reliable surrogate for the true probabilistic constraint in stochastic environments."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Introduces RAPCs and a contraction Bellman operator that jointly enforce probabilistic reach-avoid constraints while minimizing expected costs in stochastic RL, with almost-sure convergence to local optima."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Reach-avoid probability certificates turn stochastic safety constraints into a surrogate objective that reinforcement learners can optimize for minimum cost."}],"snapshot_sha256":"cb98d4fcf3484c15be6e8aa3859b7521c610e76c5361292e128e68ddb09613eb"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"b89d607eea2972b64737389308f812ef68667cb70e74931bfba336853027a791"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T11:34:51.555285Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T09:01:17.174166Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T08:02:38.043163Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.11975/integrity.json","findings":[],"snapshot_sha256":"64248693b94f85b78bf902b8a5cb22190e5b51432067cb8d1578a25192c7eb03","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study stochastic minimum-cost reach-avoid reinforcement learning, where an agent must satisfy a reach-avoid specification with probability at least $p$ while minimizing expected cumulative costs in stochastic environments. Existing safe and constrained reinforcement learning methods typically fail to jointly enforce probabilistic reach-avoid constraints and optimize cost in the learning setting in stochastic environments. To address this challenge, we introduce reach-avoid probability certificates (RAPCs), which identify states from which stochastic reach-avoid constraints are satisfiable. ","authors_text":"Bai Xue, Jingduo Pan, Taoran Wu, Yiling Xue","cross_cats":[],"headline":"Reach-avoid probability certificates turn stochastic safety constraints into a surrogate objective that reinforcement learners can optimize for minimum cost.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T11:31:36Z","title":"Stochastic Minimum-Cost Reach-Avoid Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.11975","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-13T07:23:59.690351Z","id":"dbc54e43-26df-498b-8077-da1985984ccf","model_set":{"reader":"grok-4.3"},"one_line_summary":"Introduces RAPCs and a contraction Bellman operator that jointly enforce probabilistic reach-avoid constraints while minimizing expected costs in stochastic RL, with almost-sure convergence to local optima.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Reach-avoid probability certificates turn stochastic safety constraints into a surrogate objective that reinforcement learners can optimize for minimum cost.","strongest_claim":"We establish almost sure convergence of the proposed algorithms to locally optimal policies with respect to the resulting objective.","weakest_assumption":"That reach-avoid probability certificates can be computed or approximated accurately enough during learning to serve as a reliable surrogate for the true probabilistic constraint in stochastic environments."}},"verdict_id":"dbc54e43-26df-498b-8077-da1985984ccf"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:275536e5930903903f96502298d493ad402b32f7177532778577fdbbed5002ce","target":"record","created_at":"2026-05-20T00:04:36Z","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":"0a8a6fcc79b1dafa5a3fd8818da64a1986023d846c8114d99a5698cd87cbac34","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T11:31:36Z","title_canon_sha256":"66d720cff133972aa7823b5ebaac80077487d3c568a5519b8ddfdb598a6eb03a"},"schema_version":"1.0","source":{"id":"2605.11975","kind":"arxiv","version":2}},"canonical_sha256":"02a7c7835fa949d0a47fc3c59ff73321548bc7f5ee6ec7f71478e8fe76a26526","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"02a7c7835fa949d0a47fc3c59ff73321548bc7f5ee6ec7f71478e8fe76a26526","first_computed_at":"2026-05-20T00:04:36.370095Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:04:36.370095Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WRqpCkg+0T2fhlfdtDikkWkvvI6Ioz9Sb/Yx5sMSylWGTPipgLkkFamoAHVzCaWltmYCkoT+ps+OsoNUWppXDQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:04:36.371184Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.11975","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:275536e5930903903f96502298d493ad402b32f7177532778577fdbbed5002ce","sha256:d25b7c4af974a6e6bd366395c846292f69eee536890a3eecca3ed6a7793635eb"],"state_sha256":"cc7442d764878abef4100134f6b08a9897cf25fad36a796d28384c0ad3857fa5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JoESTpzA4wRyXzuWfST+FTfyXnzQ0rHgPQqQOOzKQgN43n+7dhwCOzO++XhsuKOpK2w+ze/Yfrmdg4dKwbewCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-12T07:10:49.795609Z","bundle_sha256":"df675779c126c6b29c223972b7d4ea386c29c9b49e396479d2e75a4745932115"}}