{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:QZD6HLBJACL2DKVH23HGATGWJN","short_pith_number":"pith:QZD6HLBJ","canonical_record":{"source":{"id":"2403.03672","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-06T12:49:08Z","cross_cats_sorted":[],"title_canon_sha256":"cc13c921ccaa937c5b6c5cf25ae51e4d125e32c44fe799850e30b5f5aabd5986","abstract_canon_sha256":"cfebdcc684ddd00ebb129514d96b513948868dd90cb102f54014a465a56c2851"},"schema_version":"1.0"},"canonical_sha256":"8647e3ac290097a1aaa7d6ce604cd64b7d91be35d274f26660e2a25472019c6b","source":{"kind":"arxiv","id":"2403.03672","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.03672","created_at":"2026-07-05T10:10:39Z"},{"alias_kind":"arxiv_version","alias_value":"2403.03672v3","created_at":"2026-07-05T10:10:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.03672","created_at":"2026-07-05T10:10:39Z"},{"alias_kind":"pith_short_12","alias_value":"QZD6HLBJACL2","created_at":"2026-07-05T10:10:39Z"},{"alias_kind":"pith_short_16","alias_value":"QZD6HLBJACL2DKVH","created_at":"2026-07-05T10:10:39Z"},{"alias_kind":"pith_short_8","alias_value":"QZD6HLBJ","created_at":"2026-07-05T10:10:39Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:QZD6HLBJACL2DKVH23HGATGWJN","target":"record","payload":{"canonical_record":{"source":{"id":"2403.03672","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-06T12:49:08Z","cross_cats_sorted":[],"title_canon_sha256":"cc13c921ccaa937c5b6c5cf25ae51e4d125e32c44fe799850e30b5f5aabd5986","abstract_canon_sha256":"cfebdcc684ddd00ebb129514d96b513948868dd90cb102f54014a465a56c2851"},"schema_version":"1.0"},"canonical_sha256":"8647e3ac290097a1aaa7d6ce604cd64b7d91be35d274f26660e2a25472019c6b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:10:39.701688Z","signature_b64":"wx8wIJdIwRG/NRZNnprmOinj3kPtWjvF6rQ3LrYkweXhVuQug0H2prVeKLU6D6hLn1I1JulCgeCFr1+hIPDsCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8647e3ac290097a1aaa7d6ce604cd64b7d91be35d274f26660e2a25472019c6b","last_reissued_at":"2026-07-05T10:10:39.701251Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:10:39.701251Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.03672","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:10:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dtENmcNqoGIuyi8iWwu8J9u0Nl8fflXIMbb8hZBpBKqrrbhOuTpxELAL7+Yv6LL7ZFS/V6K3t1nBM3QuRFaDDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T00:50:56.672698Z"},"content_sha256":"4744b01c924f2e69f4bdb21cbaaf3269f234a31dfb87d3ffe7ad37fd44c0281f","schema_version":"1.0","event_id":"sha256:4744b01c924f2e69f4bdb21cbaaf3269f234a31dfb87d3ffe7ad37fd44c0281f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:QZD6HLBJACL2DKVH23HGATGWJN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning Adversarial MDPs with Stochastic Hard Constraints","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alberto Marchesi, Francesco Emanuele Stradi, Matteo Castiglioni, Nicola Gatti","submitted_at":"2024-03-06T12:49:08Z","abstract_excerpt":"We study online learning in constrained Markov decision processes (CMDPs) with adversarial losses and stochastic hard constraints, under bandit feedback. We consider three scenarios. In the first one, we address general CMDPs, where we design an algorithm attaining sublinear regret and cumulative positive constraints violation. In the second scenario, under the mild assumption that a policy strictly satisfying the constraints exists and is known to the learner, we design an algorithm that achieves sublinear regret while ensuring that constraints are satisfied at every episode with high probabi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.03672","kind":"arxiv","version":3},"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/2403.03672/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":"2026-07-05T10:10:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ASlyIGFGCpEfqJw1WzfnRTNkJNwfPMmnOAlwnPD5uqpnoVOmM0RL5plV6F3kwmdvCCeKmmcvE3Z6HwHu7AEJCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T00:50:56.673705Z"},"content_sha256":"202f1638e18d8ed15bfac1371dd8b9e535a0038930fec04508eadfe3053ba3bd","schema_version":"1.0","event_id":"sha256:202f1638e18d8ed15bfac1371dd8b9e535a0038930fec04508eadfe3053ba3bd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QZD6HLBJACL2DKVH23HGATGWJN/bundle.json","state_url":"https://pith.science/pith/QZD6HLBJACL2DKVH23HGATGWJN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QZD6HLBJACL2DKVH23HGATGWJN/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":"2026-08-18T00:50:56Z","links":{"resolver":"https://pith.science/pith/QZD6HLBJACL2DKVH23HGATGWJN","bundle":"https://pith.science/pith/QZD6HLBJACL2DKVH23HGATGWJN/bundle.json","state":"https://pith.science/pith/QZD6HLBJACL2DKVH23HGATGWJN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QZD6HLBJACL2DKVH23HGATGWJN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:QZD6HLBJACL2DKVH23HGATGWJN","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":"cfebdcc684ddd00ebb129514d96b513948868dd90cb102f54014a465a56c2851","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-06T12:49:08Z","title_canon_sha256":"cc13c921ccaa937c5b6c5cf25ae51e4d125e32c44fe799850e30b5f5aabd5986"},"schema_version":"1.0","source":{"id":"2403.03672","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.03672","created_at":"2026-07-05T10:10:39Z"},{"alias_kind":"arxiv_version","alias_value":"2403.03672v3","created_at":"2026-07-05T10:10:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.03672","created_at":"2026-07-05T10:10:39Z"},{"alias_kind":"pith_short_12","alias_value":"QZD6HLBJACL2","created_at":"2026-07-05T10:10:39Z"},{"alias_kind":"pith_short_16","alias_value":"QZD6HLBJACL2DKVH","created_at":"2026-07-05T10:10:39Z"},{"alias_kind":"pith_short_8","alias_value":"QZD6HLBJ","created_at":"2026-07-05T10:10:39Z"}],"graph_snapshots":[{"event_id":"sha256:202f1638e18d8ed15bfac1371dd8b9e535a0038930fec04508eadfe3053ba3bd","target":"graph","created_at":"2026-07-05T10:10:39Z","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/2403.03672/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study online learning in constrained Markov decision processes (CMDPs) with adversarial losses and stochastic hard constraints, under bandit feedback. We consider three scenarios. In the first one, we address general CMDPs, where we design an algorithm attaining sublinear regret and cumulative positive constraints violation. In the second scenario, under the mild assumption that a policy strictly satisfying the constraints exists and is known to the learner, we design an algorithm that achieves sublinear regret while ensuring that constraints are satisfied at every episode with high probabi","authors_text":"Alberto Marchesi, Francesco Emanuele Stradi, Matteo Castiglioni, Nicola Gatti","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-06T12:49:08Z","title":"Learning Adversarial MDPs with Stochastic Hard Constraints"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.03672","kind":"arxiv","version":3},"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":"sha256:4744b01c924f2e69f4bdb21cbaaf3269f234a31dfb87d3ffe7ad37fd44c0281f","target":"record","created_at":"2026-07-05T10:10:39Z","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":"cfebdcc684ddd00ebb129514d96b513948868dd90cb102f54014a465a56c2851","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-06T12:49:08Z","title_canon_sha256":"cc13c921ccaa937c5b6c5cf25ae51e4d125e32c44fe799850e30b5f5aabd5986"},"schema_version":"1.0","source":{"id":"2403.03672","kind":"arxiv","version":3}},"canonical_sha256":"8647e3ac290097a1aaa7d6ce604cd64b7d91be35d274f26660e2a25472019c6b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8647e3ac290097a1aaa7d6ce604cd64b7d91be35d274f26660e2a25472019c6b","first_computed_at":"2026-07-05T10:10:39.701251Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:10:39.701251Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wx8wIJdIwRG/NRZNnprmOinj3kPtWjvF6rQ3LrYkweXhVuQug0H2prVeKLU6D6hLn1I1JulCgeCFr1+hIPDsCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:10:39.701688Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.03672","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4744b01c924f2e69f4bdb21cbaaf3269f234a31dfb87d3ffe7ad37fd44c0281f","sha256:202f1638e18d8ed15bfac1371dd8b9e535a0038930fec04508eadfe3053ba3bd"],"state_sha256":"8973dcb23de01b175521dafdff195581c52ae26beff91af02f5fe3852a2c02fd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JfF1n0SOvNzkCPrsRxnJMT/DDjiDaXwFKsNjwRmf4zYQxGbdTqYJZOUtSQfgVLndPuiIZgmSboUBD3XU84vGCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T00:50:56.680702Z","bundle_sha256":"2e341df3162b7a34174ae1b23a526add478f255826b0d797c4cb7b42ee54d866"}}