{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:D2XD3IJYKETOA6N7FN2HIKIYEI","short_pith_number":"pith:D2XD3IJY","schema_version":"1.0","canonical_sha256":"1eae3da1385126e079bf2b74742918221f9c912b3a7ca35111f12e342966488f","source":{"kind":"arxiv","id":"2407.21748","version":1},"attestation_state":"computed","paper":{"title":"Diagnostic Runtime Monitoring with Martingales","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Ali Hindy, Edward Schmerling, Jonathan Kuck, Marco Pavone, Rachel Luo, Somrita Banerjee","submitted_at":"2024-07-31T17:05:10Z","abstract_excerpt":"Machine learning systems deployed in safety-critical robotics settings must be robust to distribution shifts. However, system designers must understand the cause of a distribution shift in order to implement the appropriate intervention or mitigation strategy and prevent system failure. In this paper, we present a novel framework for diagnosing distribution shifts in a streaming fashion by deploying multiple stochastic martingales simultaneously. We show that knowledge of the underlying cause of a distribution shift can lead to proper interventions over the lifecycle of a deployed system. Our "},"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":"2407.21748","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-07-31T17:05:10Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9a6fdf3bc041fcbe1aee972fae545faef434b3acca6b3d389639e55305629514","abstract_canon_sha256":"8b654858d009d55c56989ec80987a1a1bfc5cd7f7193fa9e790d902e8bcd9d35"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:50:43.202401Z","signature_b64":"daxVQt/oqLI5jHk62tOdIcCxaN0yGNS2pTG3t4fTudg84U5opluOgY2GNNP3yhLKb/3kbpswwIIKlw2RfHmIAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1eae3da1385126e079bf2b74742918221f9c912b3a7ca35111f12e342966488f","last_reissued_at":"2026-07-05T08:50:43.202006Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:50:43.202006Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diagnostic Runtime Monitoring with Martingales","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Ali Hindy, Edward Schmerling, Jonathan Kuck, Marco Pavone, Rachel Luo, Somrita Banerjee","submitted_at":"2024-07-31T17:05:10Z","abstract_excerpt":"Machine learning systems deployed in safety-critical robotics settings must be robust to distribution shifts. However, system designers must understand the cause of a distribution shift in order to implement the appropriate intervention or mitigation strategy and prevent system failure. In this paper, we present a novel framework for diagnosing distribution shifts in a streaming fashion by deploying multiple stochastic martingales simultaneously. We show that knowledge of the underlying cause of a distribution shift can lead to proper interventions over the lifecycle of a deployed system. Our "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.21748","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/2407.21748/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":"2407.21748","created_at":"2026-07-05T08:50:43.202067+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.21748v1","created_at":"2026-07-05T08:50:43.202067+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.21748","created_at":"2026-07-05T08:50:43.202067+00:00"},{"alias_kind":"pith_short_12","alias_value":"D2XD3IJYKETO","created_at":"2026-07-05T08:50:43.202067+00:00"},{"alias_kind":"pith_short_16","alias_value":"D2XD3IJYKETOA6N7","created_at":"2026-07-05T08:50:43.202067+00:00"},{"alias_kind":"pith_short_8","alias_value":"D2XD3IJY","created_at":"2026-07-05T08:50:43.202067+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.00756","citing_title":"\"Who experiences large model decay and why?\" A Hierarchical Framework for Diagnosing Heterogeneous Performance Drift","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/D2XD3IJYKETOA6N7FN2HIKIYEI","json":"https://pith.science/pith/D2XD3IJYKETOA6N7FN2HIKIYEI.json","graph_json":"https://pith.science/api/pith-number/D2XD3IJYKETOA6N7FN2HIKIYEI/graph.json","events_json":"https://pith.science/api/pith-number/D2XD3IJYKETOA6N7FN2HIKIYEI/events.json","paper":"https://pith.science/paper/D2XD3IJY"},"agent_actions":{"view_html":"https://pith.science/pith/D2XD3IJYKETOA6N7FN2HIKIYEI","download_json":"https://pith.science/pith/D2XD3IJYKETOA6N7FN2HIKIYEI.json","view_paper":"https://pith.science/paper/D2XD3IJY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.21748&json=true","fetch_graph":"https://pith.science/api/pith-number/D2XD3IJYKETOA6N7FN2HIKIYEI/graph.json","fetch_events":"https://pith.science/api/pith-number/D2XD3IJYKETOA6N7FN2HIKIYEI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/D2XD3IJYKETOA6N7FN2HIKIYEI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/D2XD3IJYKETOA6N7FN2HIKIYEI/action/storage_attestation","attest_author":"https://pith.science/pith/D2XD3IJYKETOA6N7FN2HIKIYEI/action/author_attestation","sign_citation":"https://pith.science/pith/D2XD3IJYKETOA6N7FN2HIKIYEI/action/citation_signature","submit_replication":"https://pith.science/pith/D2XD3IJYKETOA6N7FN2HIKIYEI/action/replication_record"}},"created_at":"2026-07-05T08:50:43.202067+00:00","updated_at":"2026-07-05T08:50:43.202067+00:00"}