{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PLTP73Y6R57GOGX6AADLTNGXHQ","short_pith_number":"pith:PLTP73Y6","schema_version":"1.0","canonical_sha256":"7ae6ffef1e8f7e671afe0006b9b4d73c2080eec26287c049fea79ffed2d4235e","source":{"kind":"arxiv","id":"2310.20581","version":2},"attestation_state":"computed","paper":{"title":"Stochastic Gradient Descent for Gaussian Processes Done Right","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander Terenin, Austin Tripp, Csaba Szepesv\\'ari, David Janz, Javier Antor\\'an, Jihao Andreas Lin, Jos\\'e Miguel Hern\\'andez-Lobato, Shreyas Padhy","submitted_at":"2023-10-31T16:15:13Z","abstract_excerpt":"As is well known, both sampling from the posterior and computing the mean of the posterior in Gaussian process regression reduces to solving a large linear system of equations. We study the use of stochastic gradient descent for solving this linear system, and show that when \\emph{done right} -- by which we mean using specific insights from the optimisation and kernel communities -- stochastic gradient descent is highly effective. To that end, we introduce a particularly simple \\emph{stochastic dual descent} algorithm, explain its design in an intuitive manner and illustrate the design choices"},"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":"2310.20581","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-31T16:15:13Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"cd0517924441297dad44b2465af9df75a95609a21b5adc54b43a0eea9916dd25","abstract_canon_sha256":"9c8e628edcdba9ea82ad5037c5df29dda2b1fe3c9ad2e0e6c228598d1b7307bc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:17:05.585075Z","signature_b64":"moNUoxkJxzBEnzzW/WQICsKk68gpeOJGPaDUITXPJoZ3sfbFat9w7sjAyRB+86AbUkXpOO/NzmhwyTDKNbf5AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7ae6ffef1e8f7e671afe0006b9b4d73c2080eec26287c049fea79ffed2d4235e","last_reissued_at":"2026-07-05T10:17:05.584578Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:17:05.584578Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Stochastic Gradient Descent for Gaussian Processes Done Right","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander Terenin, Austin Tripp, Csaba Szepesv\\'ari, David Janz, Javier Antor\\'an, Jihao Andreas Lin, Jos\\'e Miguel Hern\\'andez-Lobato, Shreyas Padhy","submitted_at":"2023-10-31T16:15:13Z","abstract_excerpt":"As is well known, both sampling from the posterior and computing the mean of the posterior in Gaussian process regression reduces to solving a large linear system of equations. We study the use of stochastic gradient descent for solving this linear system, and show that when \\emph{done right} -- by which we mean using specific insights from the optimisation and kernel communities -- stochastic gradient descent is highly effective. To that end, we introduce a particularly simple \\emph{stochastic dual descent} algorithm, explain its design in an intuitive manner and illustrate the design choices"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.20581","kind":"arxiv","version":2},"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/2310.20581/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":"2310.20581","created_at":"2026-07-05T10:17:05.584640+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.20581v2","created_at":"2026-07-05T10:17:05.584640+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.20581","created_at":"2026-07-05T10:17:05.584640+00:00"},{"alias_kind":"pith_short_12","alias_value":"PLTP73Y6R57G","created_at":"2026-07-05T10:17:05.584640+00:00"},{"alias_kind":"pith_short_16","alias_value":"PLTP73Y6R57GOGX6","created_at":"2026-07-05T10:17:05.584640+00:00"},{"alias_kind":"pith_short_8","alias_value":"PLTP73Y6","created_at":"2026-07-05T10:17:05.584640+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.15906","citing_title":"From Local Interactions to Global Operators: Scalable Gaussian Process Operator for Physical Systems","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PLTP73Y6R57GOGX6AADLTNGXHQ","json":"https://pith.science/pith/PLTP73Y6R57GOGX6AADLTNGXHQ.json","graph_json":"https://pith.science/api/pith-number/PLTP73Y6R57GOGX6AADLTNGXHQ/graph.json","events_json":"https://pith.science/api/pith-number/PLTP73Y6R57GOGX6AADLTNGXHQ/events.json","paper":"https://pith.science/paper/PLTP73Y6"},"agent_actions":{"view_html":"https://pith.science/pith/PLTP73Y6R57GOGX6AADLTNGXHQ","download_json":"https://pith.science/pith/PLTP73Y6R57GOGX6AADLTNGXHQ.json","view_paper":"https://pith.science/paper/PLTP73Y6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.20581&json=true","fetch_graph":"https://pith.science/api/pith-number/PLTP73Y6R57GOGX6AADLTNGXHQ/graph.json","fetch_events":"https://pith.science/api/pith-number/PLTP73Y6R57GOGX6AADLTNGXHQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PLTP73Y6R57GOGX6AADLTNGXHQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PLTP73Y6R57GOGX6AADLTNGXHQ/action/storage_attestation","attest_author":"https://pith.science/pith/PLTP73Y6R57GOGX6AADLTNGXHQ/action/author_attestation","sign_citation":"https://pith.science/pith/PLTP73Y6R57GOGX6AADLTNGXHQ/action/citation_signature","submit_replication":"https://pith.science/pith/PLTP73Y6R57GOGX6AADLTNGXHQ/action/replication_record"}},"created_at":"2026-07-05T10:17:05.584640+00:00","updated_at":"2026-07-05T10:17:05.584640+00:00"}