{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RPQF4QTCWEH53OSKWIJ635DUAJ","short_pith_number":"pith:RPQF4QTC","schema_version":"1.0","canonical_sha256":"8be05e4262b10fddba4ab213edf474024b13a2ae3a5f883eab960d66f74d02e2","source":{"kind":"arxiv","id":"2404.14758","version":1},"attestation_state":"computed","paper":{"title":"Second-order Information Promotes Mini-Batch Robustness in Variance-Reduced Gradients","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Albert S. Berahas, Micha{\\l} Derezi\\'nski, Sachin Garg","submitted_at":"2024-04-23T05:45:52Z","abstract_excerpt":"We show that, for finite-sum minimization problems, incorporating partial second-order information of the objective function can dramatically improve the robustness to mini-batch size of variance-reduced stochastic gradient methods, making them more scalable while retaining their benefits over traditional Newton-type approaches. We demonstrate this phenomenon on a prototypical stochastic second-order algorithm, called Mini-Batch Stochastic Variance-Reduced Newton ($\\texttt{Mb-SVRN}$), which combines variance-reduced gradient estimates with access to an approximate Hessian oracle. In particular"},"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":"2404.14758","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"math.OC","submitted_at":"2024-04-23T05:45:52Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"e4d427d63df46949dc344df631d87584d48b2ecbad215661b14ce0278af24bc4","abstract_canon_sha256":"751df28df71954dea3679420e20693dc9b4e80f93d76bd164fcdace39de2d6e1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:11:09.811258Z","signature_b64":"c0GNd6GpWw1ck/knww31aJ4F1hdbbVDE7RN4uL3KE6QilSZjFBnPSmJgktnj/fj/qbIMj6cNQwNanXP3stBuAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8be05e4262b10fddba4ab213edf474024b13a2ae3a5f883eab960d66f74d02e2","last_reissued_at":"2026-07-05T08:11:09.810680Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:11:09.810680Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Second-order Information Promotes Mini-Batch Robustness in Variance-Reduced Gradients","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Albert S. Berahas, Micha{\\l} Derezi\\'nski, Sachin Garg","submitted_at":"2024-04-23T05:45:52Z","abstract_excerpt":"We show that, for finite-sum minimization problems, incorporating partial second-order information of the objective function can dramatically improve the robustness to mini-batch size of variance-reduced stochastic gradient methods, making them more scalable while retaining their benefits over traditional Newton-type approaches. We demonstrate this phenomenon on a prototypical stochastic second-order algorithm, called Mini-Batch Stochastic Variance-Reduced Newton ($\\texttt{Mb-SVRN}$), which combines variance-reduced gradient estimates with access to an approximate Hessian oracle. In particular"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.14758","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/2404.14758/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":"2404.14758","created_at":"2026-07-05T08:11:09.810746+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.14758v1","created_at":"2026-07-05T08:11:09.810746+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.14758","created_at":"2026-07-05T08:11:09.810746+00:00"},{"alias_kind":"pith_short_12","alias_value":"RPQF4QTCWEH5","created_at":"2026-07-05T08:11:09.810746+00:00"},{"alias_kind":"pith_short_16","alias_value":"RPQF4QTCWEH53OSK","created_at":"2026-07-05T08:11:09.810746+00:00"},{"alias_kind":"pith_short_8","alias_value":"RPQF4QTC","created_at":"2026-07-05T08:11:09.810746+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.15941","citing_title":"SAPPHIRE: Preconditioned Stochastic Variance Reduction for Faster Large-Scale Statistical Learning","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RPQF4QTCWEH53OSKWIJ635DUAJ","json":"https://pith.science/pith/RPQF4QTCWEH53OSKWIJ635DUAJ.json","graph_json":"https://pith.science/api/pith-number/RPQF4QTCWEH53OSKWIJ635DUAJ/graph.json","events_json":"https://pith.science/api/pith-number/RPQF4QTCWEH53OSKWIJ635DUAJ/events.json","paper":"https://pith.science/paper/RPQF4QTC"},"agent_actions":{"view_html":"https://pith.science/pith/RPQF4QTCWEH53OSKWIJ635DUAJ","download_json":"https://pith.science/pith/RPQF4QTCWEH53OSKWIJ635DUAJ.json","view_paper":"https://pith.science/paper/RPQF4QTC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.14758&json=true","fetch_graph":"https://pith.science/api/pith-number/RPQF4QTCWEH53OSKWIJ635DUAJ/graph.json","fetch_events":"https://pith.science/api/pith-number/RPQF4QTCWEH53OSKWIJ635DUAJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RPQF4QTCWEH53OSKWIJ635DUAJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RPQF4QTCWEH53OSKWIJ635DUAJ/action/storage_attestation","attest_author":"https://pith.science/pith/RPQF4QTCWEH53OSKWIJ635DUAJ/action/author_attestation","sign_citation":"https://pith.science/pith/RPQF4QTCWEH53OSKWIJ635DUAJ/action/citation_signature","submit_replication":"https://pith.science/pith/RPQF4QTCWEH53OSKWIJ635DUAJ/action/replication_record"}},"created_at":"2026-07-05T08:11:09.810746+00:00","updated_at":"2026-07-05T08:11:09.810746+00:00"}