{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:AXFLYWSAPDRHXZ533RPH6JAJ42","short_pith_number":"pith:AXFLYWSA","canonical_record":{"source":{"id":"2505.21841","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-28T00:16:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"669ebc57ffe93d05b66ea3d5c5074d0e430e99d83aa00db769832b0e076385d6","abstract_canon_sha256":"425b9288ff5e3dfebef159914bf4a0cb35d9d70ab557c34498f271b40e7067d6"},"schema_version":"1.0"},"canonical_sha256":"05cabc5a4078e27be7bbdc5e7f2409e6adbae9b81a15daab05ce3aa06eb10c82","source":{"kind":"arxiv","id":"2505.21841","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.21841","created_at":"2026-07-05T11:13:27Z"},{"alias_kind":"arxiv_version","alias_value":"2505.21841v1","created_at":"2026-07-05T11:13:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21841","created_at":"2026-07-05T11:13:27Z"},{"alias_kind":"pith_short_12","alias_value":"AXFLYWSAPDRH","created_at":"2026-07-05T11:13:27Z"},{"alias_kind":"pith_short_16","alias_value":"AXFLYWSAPDRHXZ53","created_at":"2026-07-05T11:13:27Z"},{"alias_kind":"pith_short_8","alias_value":"AXFLYWSA","created_at":"2026-07-05T11:13:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:AXFLYWSAPDRHXZ533RPH6JAJ42","target":"record","payload":{"canonical_record":{"source":{"id":"2505.21841","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-28T00:16:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"669ebc57ffe93d05b66ea3d5c5074d0e430e99d83aa00db769832b0e076385d6","abstract_canon_sha256":"425b9288ff5e3dfebef159914bf4a0cb35d9d70ab557c34498f271b40e7067d6"},"schema_version":"1.0"},"canonical_sha256":"05cabc5a4078e27be7bbdc5e7f2409e6adbae9b81a15daab05ce3aa06eb10c82","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:27.218112Z","signature_b64":"AsA+TtsspB71QcM2FjZx1aDPyvCw8IqMYWDWQjm2sPQgbwQPO+roMkGFkC93xWVEzldKFnUkdEQVcCLkvGSlCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"05cabc5a4078e27be7bbdc5e7f2409e6adbae9b81a15daab05ce3aa06eb10c82","last_reissued_at":"2026-07-05T11:13:27.217616Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:27.217616Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.21841","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":"2026-07-05T11:13:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xmLTy/f/nRnYBdf9hNLJfHAIv7K4Wv65tNwGRK7le063+AxMS6//G3p+FnDTPY/BH8Xtzqbe5F4Z96SBoO2/CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T01:01:31.842562Z"},"content_sha256":"ee32702569f810645bc871e1ef115c50055ba835861dc91b277fbc51ee016335","schema_version":"1.0","event_id":"sha256:ee32702569f810645bc871e1ef115c50055ba835861dc91b277fbc51ee016335"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:AXFLYWSAPDRHXZ533RPH6JAJ42","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"An Optimistic Algorithm for online CMDPS with Anytime Adversarial Constraints","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dabeen Lee, Honghao Wei, Jiahui Zhu, Kihyun Yu, Xin Liu","submitted_at":"2025-05-28T00:16:34Z","abstract_excerpt":"Online safe reinforcement learning (RL) plays a key role in dynamic environments, with applications in autonomous driving, robotics, and cybersecurity. The objective is to learn optimal policies that maximize rewards while satisfying safety constraints modeled by constrained Markov decision processes (CMDPs). Existing methods achieve sublinear regret under stochastic constraints but often fail in adversarial settings, where constraints are unknown, time-varying, and potentially adversarially designed. In this paper, we propose the Optimistic Mirror Descent Primal-Dual (OMDPD) algorithm, the fi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21841","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/2505.21841/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":"2026-07-05T11:13:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6FGCd81SriiR3zdj96/OJhcor4zLqPSgd9RWqdf33SkwgYSKNiuDNTh1hbGzHlnuU0PIuclaOAXw7V3rMF28CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T01:01:31.843480Z"},"content_sha256":"9dfba194acfc10ffd99f4731f603973f553c330765efd118076588dae9c581e8","schema_version":"1.0","event_id":"sha256:9dfba194acfc10ffd99f4731f603973f553c330765efd118076588dae9c581e8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AXFLYWSAPDRHXZ533RPH6JAJ42/bundle.json","state_url":"https://pith.science/pith/AXFLYWSAPDRHXZ533RPH6JAJ42/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AXFLYWSAPDRHXZ533RPH6JAJ42/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":"2026-08-08T01:01:31Z","links":{"resolver":"https://pith.science/pith/AXFLYWSAPDRHXZ533RPH6JAJ42","bundle":"https://pith.science/pith/AXFLYWSAPDRHXZ533RPH6JAJ42/bundle.json","state":"https://pith.science/pith/AXFLYWSAPDRHXZ533RPH6JAJ42/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AXFLYWSAPDRHXZ533RPH6JAJ42/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:AXFLYWSAPDRHXZ533RPH6JAJ42","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":"425b9288ff5e3dfebef159914bf4a0cb35d9d70ab557c34498f271b40e7067d6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-28T00:16:34Z","title_canon_sha256":"669ebc57ffe93d05b66ea3d5c5074d0e430e99d83aa00db769832b0e076385d6"},"schema_version":"1.0","source":{"id":"2505.21841","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.21841","created_at":"2026-07-05T11:13:27Z"},{"alias_kind":"arxiv_version","alias_value":"2505.21841v1","created_at":"2026-07-05T11:13:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21841","created_at":"2026-07-05T11:13:27Z"},{"alias_kind":"pith_short_12","alias_value":"AXFLYWSAPDRH","created_at":"2026-07-05T11:13:27Z"},{"alias_kind":"pith_short_16","alias_value":"AXFLYWSAPDRHXZ53","created_at":"2026-07-05T11:13:27Z"},{"alias_kind":"pith_short_8","alias_value":"AXFLYWSA","created_at":"2026-07-05T11:13:27Z"}],"graph_snapshots":[{"event_id":"sha256:9dfba194acfc10ffd99f4731f603973f553c330765efd118076588dae9c581e8","target":"graph","created_at":"2026-07-05T11:13:27Z","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/2505.21841/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Online safe reinforcement learning (RL) plays a key role in dynamic environments, with applications in autonomous driving, robotics, and cybersecurity. The objective is to learn optimal policies that maximize rewards while satisfying safety constraints modeled by constrained Markov decision processes (CMDPs). Existing methods achieve sublinear regret under stochastic constraints but often fail in adversarial settings, where constraints are unknown, time-varying, and potentially adversarially designed. In this paper, we propose the Optimistic Mirror Descent Primal-Dual (OMDPD) algorithm, the fi","authors_text":"Dabeen Lee, Honghao Wei, Jiahui Zhu, Kihyun Yu, Xin Liu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-28T00:16:34Z","title":"An Optimistic Algorithm for online CMDPS with Anytime Adversarial Constraints"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21841","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:ee32702569f810645bc871e1ef115c50055ba835861dc91b277fbc51ee016335","target":"record","created_at":"2026-07-05T11:13:27Z","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":"425b9288ff5e3dfebef159914bf4a0cb35d9d70ab557c34498f271b40e7067d6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-28T00:16:34Z","title_canon_sha256":"669ebc57ffe93d05b66ea3d5c5074d0e430e99d83aa00db769832b0e076385d6"},"schema_version":"1.0","source":{"id":"2505.21841","kind":"arxiv","version":1}},"canonical_sha256":"05cabc5a4078e27be7bbdc5e7f2409e6adbae9b81a15daab05ce3aa06eb10c82","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"05cabc5a4078e27be7bbdc5e7f2409e6adbae9b81a15daab05ce3aa06eb10c82","first_computed_at":"2026-07-05T11:13:27.217616Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:27.217616Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"AsA+TtsspB71QcM2FjZx1aDPyvCw8IqMYWDWQjm2sPQgbwQPO+roMkGFkC93xWVEzldKFnUkdEQVcCLkvGSlCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:27.218112Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.21841","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ee32702569f810645bc871e1ef115c50055ba835861dc91b277fbc51ee016335","sha256:9dfba194acfc10ffd99f4731f603973f553c330765efd118076588dae9c581e8"],"state_sha256":"8016d6ad5270d002221c1b69cb3c8696487c5ac1be12aa081292f7d147207cda"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lrj0m1GYjYCrtw7If7+8UQkWShJux/j/uy40jbZGMA1XHqQSBc//WuF+TGf34TByFiHvaN8QCk9jNe3vnsg/CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T01:01:31.850194Z","bundle_sha256":"d371ffcdcb31f789e76b4425521b5904845d0175b817676270b3470379ef9b77"}}