{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:YH3G5FET4ABA3LU3HGQRJAB6EP","short_pith_number":"pith:YH3G5FET","schema_version":"1.0","canonical_sha256":"c1f66e9493e0020dae9b39a114803e23ea37396f8b514a29780d17c8cdd5af60","source":{"kind":"arxiv","id":"2305.19535","version":3},"attestation_state":"computed","paper":{"title":"Low-rank extended Kalman filtering for online learning of neural networks from streaming data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alexander Y Shestopaloff, Gerardo Dur\\'an-Mart\\'in, Kevin Murphy, Matt Jones, Peter G. Chang","submitted_at":"2023-05-31T03:48:49Z","abstract_excerpt":"We propose an efficient online approximate Bayesian inference algorithm for estimating the parameters of a nonlinear function from a potentially non-stationary data stream. The method is based on the extended Kalman filter (EKF), but uses a novel low-rank plus diagonal decomposition of the posterior precision matrix, which gives a cost per step which is linear in the number of model parameters. In contrast to methods based on stochastic variational inference, our method is fully deterministic, and does not require step-size tuning. We show experimentally that this results in much faster (more "},"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":"2305.19535","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-05-31T03:48:49Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b514578263ad60734fb8a3ede978f8411ae6bbfb3bf3046051844358cd83e938","abstract_canon_sha256":"c602d7525590c869800c5be94a60ae141b09affe3e4952cce6a0dbd8c3389ec6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:25:41.814779Z","signature_b64":"P4afBJ7iygSrC7O/kIDuJUouk4WB6kohmHLFp9ASw+E+Rq15feGOmFNicOf9jwYTj061/elJuUmROQKLPp4GAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c1f66e9493e0020dae9b39a114803e23ea37396f8b514a29780d17c8cdd5af60","last_reissued_at":"2026-07-05T06:25:41.814308Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:25:41.814308Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Low-rank extended Kalman filtering for online learning of neural networks from streaming data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alexander Y Shestopaloff, Gerardo Dur\\'an-Mart\\'in, Kevin Murphy, Matt Jones, Peter G. Chang","submitted_at":"2023-05-31T03:48:49Z","abstract_excerpt":"We propose an efficient online approximate Bayesian inference algorithm for estimating the parameters of a nonlinear function from a potentially non-stationary data stream. The method is based on the extended Kalman filter (EKF), but uses a novel low-rank plus diagonal decomposition of the posterior precision matrix, which gives a cost per step which is linear in the number of model parameters. In contrast to methods based on stochastic variational inference, our method is fully deterministic, and does not require step-size tuning. We show experimentally that this results in much faster (more "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.19535","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/2305.19535/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":"2305.19535","created_at":"2026-07-05T06:25:41.814377+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.19535v3","created_at":"2026-07-05T06:25:41.814377+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.19535","created_at":"2026-07-05T06:25:41.814377+00:00"},{"alias_kind":"pith_short_12","alias_value":"YH3G5FET4ABA","created_at":"2026-07-05T06:25:41.814377+00:00"},{"alias_kind":"pith_short_16","alias_value":"YH3G5FET4ABA3LU3","created_at":"2026-07-05T06:25:41.814377+00:00"},{"alias_kind":"pith_short_8","alias_value":"YH3G5FET","created_at":"2026-07-05T06:25:41.814377+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/YH3G5FET4ABA3LU3HGQRJAB6EP","json":"https://pith.science/pith/YH3G5FET4ABA3LU3HGQRJAB6EP.json","graph_json":"https://pith.science/api/pith-number/YH3G5FET4ABA3LU3HGQRJAB6EP/graph.json","events_json":"https://pith.science/api/pith-number/YH3G5FET4ABA3LU3HGQRJAB6EP/events.json","paper":"https://pith.science/paper/YH3G5FET"},"agent_actions":{"view_html":"https://pith.science/pith/YH3G5FET4ABA3LU3HGQRJAB6EP","download_json":"https://pith.science/pith/YH3G5FET4ABA3LU3HGQRJAB6EP.json","view_paper":"https://pith.science/paper/YH3G5FET","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.19535&json=true","fetch_graph":"https://pith.science/api/pith-number/YH3G5FET4ABA3LU3HGQRJAB6EP/graph.json","fetch_events":"https://pith.science/api/pith-number/YH3G5FET4ABA3LU3HGQRJAB6EP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YH3G5FET4ABA3LU3HGQRJAB6EP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YH3G5FET4ABA3LU3HGQRJAB6EP/action/storage_attestation","attest_author":"https://pith.science/pith/YH3G5FET4ABA3LU3HGQRJAB6EP/action/author_attestation","sign_citation":"https://pith.science/pith/YH3G5FET4ABA3LU3HGQRJAB6EP/action/citation_signature","submit_replication":"https://pith.science/pith/YH3G5FET4ABA3LU3HGQRJAB6EP/action/replication_record"}},"created_at":"2026-07-05T06:25:41.814377+00:00","updated_at":"2026-07-05T06:25:41.814377+00:00"}