{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DY6SHWTIE2HRRVK5RWVBUPCEKU","short_pith_number":"pith:DY6SHWTI","schema_version":"1.0","canonical_sha256":"1e3d23da68268f18d55d8daa1a3c4455000d988e06f2fee1f35a815bc8b1e2ce","source":{"kind":"arxiv","id":"2607.15457","version":1},"attestation_state":"computed","paper":{"title":"Robust Peak-cost Constrained Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Arnob Ghosh, Debdipta Goswami, Honghao Wei, Santosh Mohan Rajkumar, Shilpa Mukhopadhyay, Sourav Ganguly","submitted_at":"2026-07-16T21:00:50Z","abstract_excerpt":"We study robust peak-cost constrained reinforcement learning (RP-CRL), where the objective is to maximize expected reward while controlling the maximum cost encountered along a trajectory. This setting is motivated by safety-critical applications in which a single large violation can be catastrophic and therefore cannot be adequately captured by the standard CMDP framework based on expected cumulative cost. Existing reachability-constrained RL methods adopt Lagrangian-based approaches, yet the underlying duality properties of peak-cost constrained MDPs remain unclear. We show that, unlike stan"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.15457","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-16T21:00:50Z","cross_cats_sorted":[],"title_canon_sha256":"7c0208ccdc562ef175b837dcf6114e7c797df32c674252da4d3312287aafd278","abstract_canon_sha256":"d13795186ec9628d766c4b6e188a5ba3eca5bc73a3154ddc4dfdc1e1b2176d58"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-20T00:18:35.022931Z","signature_b64":"p28aVYtkEmduJavQl6wbHLv/BZy4k/fkrXC1zuP7pL62KqDO/1lUJykiQrXbtvkxbZ6FYRWdWKUEqCRf0I9hDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e3d23da68268f18d55d8daa1a3c4455000d988e06f2fee1f35a815bc8b1e2ce","last_reissued_at":"2026-07-20T00:18:35.022083Z","signature_status":"signed_v1","first_computed_at":"2026-07-20T00:18:35.022083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Robust Peak-cost Constrained Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Arnob Ghosh, Debdipta Goswami, Honghao Wei, Santosh Mohan Rajkumar, Shilpa Mukhopadhyay, Sourav Ganguly","submitted_at":"2026-07-16T21:00:50Z","abstract_excerpt":"We study robust peak-cost constrained reinforcement learning (RP-CRL), where the objective is to maximize expected reward while controlling the maximum cost encountered along a trajectory. This setting is motivated by safety-critical applications in which a single large violation can be catastrophic and therefore cannot be adequately captured by the standard CMDP framework based on expected cumulative cost. Existing reachability-constrained RL methods adopt Lagrangian-based approaches, yet the underlying duality properties of peak-cost constrained MDPs remain unclear. We show that, unlike stan"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.15457","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/2607.15457/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.15457","created_at":"2026-07-20T00:18:35.022518+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.15457v1","created_at":"2026-07-20T00:18:35.022518+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.15457","created_at":"2026-07-20T00:18:35.022518+00:00"},{"alias_kind":"pith_short_12","alias_value":"DY6SHWTIE2HR","created_at":"2026-07-20T00:18:35.022518+00:00"},{"alias_kind":"pith_short_16","alias_value":"DY6SHWTIE2HRRVK5","created_at":"2026-07-20T00:18:35.022518+00:00"},{"alias_kind":"pith_short_8","alias_value":"DY6SHWTI","created_at":"2026-07-20T00:18:35.022518+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DY6SHWTIE2HRRVK5RWVBUPCEKU","json":"https://pith.science/pith/DY6SHWTIE2HRRVK5RWVBUPCEKU.json","graph_json":"https://pith.science/api/pith-number/DY6SHWTIE2HRRVK5RWVBUPCEKU/graph.json","events_json":"https://pith.science/api/pith-number/DY6SHWTIE2HRRVK5RWVBUPCEKU/events.json","paper":"https://pith.science/paper/DY6SHWTI"},"agent_actions":{"view_html":"https://pith.science/pith/DY6SHWTIE2HRRVK5RWVBUPCEKU","download_json":"https://pith.science/pith/DY6SHWTIE2HRRVK5RWVBUPCEKU.json","view_paper":"https://pith.science/paper/DY6SHWTI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.15457&json=true","fetch_graph":"https://pith.science/api/pith-number/DY6SHWTIE2HRRVK5RWVBUPCEKU/graph.json","fetch_events":"https://pith.science/api/pith-number/DY6SHWTIE2HRRVK5RWVBUPCEKU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DY6SHWTIE2HRRVK5RWVBUPCEKU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DY6SHWTIE2HRRVK5RWVBUPCEKU/action/storage_attestation","attest_author":"https://pith.science/pith/DY6SHWTIE2HRRVK5RWVBUPCEKU/action/author_attestation","sign_citation":"https://pith.science/pith/DY6SHWTIE2HRRVK5RWVBUPCEKU/action/citation_signature","submit_replication":"https://pith.science/pith/DY6SHWTIE2HRRVK5RWVBUPCEKU/action/replication_record"}},"created_at":"2026-07-20T00:18:35.022518+00:00","updated_at":"2026-07-20T00:18:35.022518+00:00"}