{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:A6OWQGJDYPS4J77SQY25SC2XFT","short_pith_number":"pith:A6OWQGJD","schema_version":"1.0","canonical_sha256":"079d681923c3e5c4fff28635d90b572cf4b0c5439b3698725343a67157ed08e4","source":{"kind":"arxiv","id":"2601.02527","version":2},"attestation_state":"computed","paper":{"title":"Scalable Gaussian Processes for Integrated and Overlapping Measurements Via Augmented State Space Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Augmenting state space models with a resetting integral state yields the exact Gaussian process posterior for integrated and overlapping measurements in linear time.","cross_cats":["astro-ph.EP","astro-ph.SR"],"primary_cat":"astro-ph.IM","authors_text":"Daniel Foreman-Mackey, Jacob K. Luhn, Megan Bedell, Ryan A. Rubenzahl, Simo S\\\"arkk\\\"a, Soichiro Hattori, Will M. Farr","submitted_at":"2026-01-05T20:00:50Z","abstract_excerpt":"Astronomical measurements are often integrated over finite exposures, which can obscure latent variability on comparable timescales. Correctly accounting for exposure integration with Gaussian Processes (GPs) in such scenarios is essential but computationally challenging: once exposure times vary or overlap across measurements, the covariance matrix forfeits any quasiseparability, forcing O($N^2$) memory and O($N^3$) runtime costs. Linear Gaussian state space models (SSMs) are equivalent to GPs and have well-known O($N$) solutions via the Kalman filter and RTS smoother. In this work, we extend"},"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":"2601.02527","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"astro-ph.IM","submitted_at":"2026-01-05T20:00:50Z","cross_cats_sorted":["astro-ph.EP","astro-ph.SR"],"title_canon_sha256":"3ec231ef3c4846144fa320a7a54521baed33f5f79424e345152d2be10ee5aaa9","abstract_canon_sha256":"cf83ed37291ae3b7ff4ed11926cdd982154375bd5b33ccfa29b69a39411d52f7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:10:11.414759Z","signature_b64":"9luSS9KkkH3SppYhAUc5NTB8qgxdpw2w2PIHXfdmspVjbEeCnqUVlPh654BoJ4W44zpp3x6icqHTN5kGS1KLBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"079d681923c3e5c4fff28635d90b572cf4b0c5439b3698725343a67157ed08e4","last_reissued_at":"2026-05-18T03:10:11.414205Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:10:11.414205Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scalable Gaussian Processes for Integrated and Overlapping Measurements Via Augmented State Space Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Augmenting state space models with a resetting integral state yields the exact Gaussian process posterior for integrated and overlapping measurements in linear time.","cross_cats":["astro-ph.EP","astro-ph.SR"],"primary_cat":"astro-ph.IM","authors_text":"Daniel Foreman-Mackey, Jacob K. Luhn, Megan Bedell, Ryan A. Rubenzahl, Simo S\\\"arkk\\\"a, Soichiro Hattori, Will M. Farr","submitted_at":"2026-01-05T20:00:50Z","abstract_excerpt":"Astronomical measurements are often integrated over finite exposures, which can obscure latent variability on comparable timescales. Correctly accounting for exposure integration with Gaussian Processes (GPs) in such scenarios is essential but computationally challenging: once exposure times vary or overlap across measurements, the covariance matrix forfeits any quasiseparability, forcing O($N^2$) memory and O($N^3$) runtime costs. Linear Gaussian state space models (SSMs) are equivalent to GPs and have well-known O($N$) solutions via the Kalman filter and RTS smoother. In this work, we extend"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"This construction yields exactly the same posterior as a fully integrated GP but in O(N) time on a CPU, and is parallelizable down to O(N/T + log T) on a GPU with T parallel workers.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The augmentation with the integral state exactly reproduces the integrated measurement covariance without approximation or loss of the GP equivalence, which must hold for arbitrary exposure start and end times.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Augmented state space models allow exact Gaussian process posteriors for integrated and overlapping measurements in O(N) time.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Augmenting state space models with a resetting integral state yields the exact Gaussian process posterior for integrated and overlapping measurements in linear time.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"c8d907fc4abf6f9e9b52777e6ce1116e7639bbeaa23a982ed8aa9fcc9b0d6a8c"},"source":{"id":"2601.02527","kind":"arxiv","version":2},"verdict":{"id":"b47fab90-b8e8-4c1b-8e41-abbd6381c4b7","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T17:15:03.768184Z","strongest_claim":"This construction yields exactly the same posterior as a fully integrated GP but in O(N) time on a CPU, and is parallelizable down to O(N/T + log T) on a GPU with T parallel workers.","one_line_summary":"Augmented state space models allow exact Gaussian process posteriors for integrated and overlapping measurements in O(N) time.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The augmentation with the integral state exactly reproduces the integrated measurement covariance without approximation or loss of the GP equivalence, which must hold for arbitrary exposure start and end times.","pith_extraction_headline":"Augmenting state space models with a resetting integral state yields the exact Gaussian process posterior for integrated and overlapping measurements in linear time."},"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":"2601.02527","created_at":"2026-05-18T03:10:11.414303+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.02527v2","created_at":"2026-05-18T03:10:11.414303+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.02527","created_at":"2026-05-18T03:10:11.414303+00:00"},{"alias_kind":"pith_short_12","alias_value":"A6OWQGJDYPS4","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"A6OWQGJDYPS4J77S","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"A6OWQGJD","created_at":"2026-05-18T12:33:37.589309+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/A6OWQGJDYPS4J77SQY25SC2XFT","json":"https://pith.science/pith/A6OWQGJDYPS4J77SQY25SC2XFT.json","graph_json":"https://pith.science/api/pith-number/A6OWQGJDYPS4J77SQY25SC2XFT/graph.json","events_json":"https://pith.science/api/pith-number/A6OWQGJDYPS4J77SQY25SC2XFT/events.json","paper":"https://pith.science/paper/A6OWQGJD"},"agent_actions":{"view_html":"https://pith.science/pith/A6OWQGJDYPS4J77SQY25SC2XFT","download_json":"https://pith.science/pith/A6OWQGJDYPS4J77SQY25SC2XFT.json","view_paper":"https://pith.science/paper/A6OWQGJD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.02527&json=true","fetch_graph":"https://pith.science/api/pith-number/A6OWQGJDYPS4J77SQY25SC2XFT/graph.json","fetch_events":"https://pith.science/api/pith-number/A6OWQGJDYPS4J77SQY25SC2XFT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A6OWQGJDYPS4J77SQY25SC2XFT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A6OWQGJDYPS4J77SQY25SC2XFT/action/storage_attestation","attest_author":"https://pith.science/pith/A6OWQGJDYPS4J77SQY25SC2XFT/action/author_attestation","sign_citation":"https://pith.science/pith/A6OWQGJDYPS4J77SQY25SC2XFT/action/citation_signature","submit_replication":"https://pith.science/pith/A6OWQGJDYPS4J77SQY25SC2XFT/action/replication_record"}},"created_at":"2026-05-18T03:10:11.414303+00:00","updated_at":"2026-05-18T03:10:11.414303+00:00"}