{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CCEILMG7APRY7PV5ESNLG2WN4I","short_pith_number":"pith:CCEILMG7","schema_version":"1.0","canonical_sha256":"108885b0df03e38fbebd249ab36acde2322c33e38a4c6483ce4efb40fd80fdce","source":{"kind":"arxiv","id":"2506.08360","version":1},"attestation_state":"computed","paper":{"title":"NysAct: A Scalable Preconditioned Gradient Descent using Nystrom Approximation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Hyunseok Seung, Hyunsuk Ko, Jaewoo Lee","submitted_at":"2025-06-10T02:18:08Z","abstract_excerpt":"Adaptive gradient methods are computationally efficient and converge quickly, but they often suffer from poor generalization. In contrast, second-order methods enhance convergence and generalization but typically incur high computational and memory costs. In this work, we introduce NysAct, a scalable first-order gradient preconditioning method that strikes a balance between state-of-the-art first-order and second-order optimization methods. NysAct leverages an eigenvalue-shifted Nystrom method to approximate the activation covariance matrix, which is used as a preconditioning matrix, significa"},"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":"2506.08360","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T02:18:08Z","cross_cats_sorted":[],"title_canon_sha256":"6dd6b262c02127e7a3744009a29f34edcdd8dfeacb1bee24ca5fb801f690a62c","abstract_canon_sha256":"091772a8e45227dc7b77f2b05d619c0dc122d3e13ee39a94b37e53a686fcb15d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:48.502989Z","signature_b64":"TFAbvgRyfgzpBks73Mkwu81/ZvOxvd7AGPk9nAiFmSucyRNAwBgU+QnDZc7cF0MajPxRcgGg6G1fMGE8mgDUAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"108885b0df03e38fbebd249ab36acde2322c33e38a4c6483ce4efb40fd80fdce","last_reissued_at":"2026-07-05T11:18:48.502582Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:48.502582Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"NysAct: A Scalable Preconditioned Gradient Descent using Nystrom Approximation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Hyunseok Seung, Hyunsuk Ko, Jaewoo Lee","submitted_at":"2025-06-10T02:18:08Z","abstract_excerpt":"Adaptive gradient methods are computationally efficient and converge quickly, but they often suffer from poor generalization. In contrast, second-order methods enhance convergence and generalization but typically incur high computational and memory costs. In this work, we introduce NysAct, a scalable first-order gradient preconditioning method that strikes a balance between state-of-the-art first-order and second-order optimization methods. NysAct leverages an eigenvalue-shifted Nystrom method to approximate the activation covariance matrix, which is used as a preconditioning matrix, significa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.08360","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/2506.08360/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":"2506.08360","created_at":"2026-07-05T11:18:48.502640+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.08360v1","created_at":"2026-07-05T11:18:48.502640+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.08360","created_at":"2026-07-05T11:18:48.502640+00:00"},{"alias_kind":"pith_short_12","alias_value":"CCEILMG7APRY","created_at":"2026-07-05T11:18:48.502640+00:00"},{"alias_kind":"pith_short_16","alias_value":"CCEILMG7APRY7PV5","created_at":"2026-07-05T11:18:48.502640+00:00"},{"alias_kind":"pith_short_8","alias_value":"CCEILMG7","created_at":"2026-07-05T11:18:48.502640+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/CCEILMG7APRY7PV5ESNLG2WN4I","json":"https://pith.science/pith/CCEILMG7APRY7PV5ESNLG2WN4I.json","graph_json":"https://pith.science/api/pith-number/CCEILMG7APRY7PV5ESNLG2WN4I/graph.json","events_json":"https://pith.science/api/pith-number/CCEILMG7APRY7PV5ESNLG2WN4I/events.json","paper":"https://pith.science/paper/CCEILMG7"},"agent_actions":{"view_html":"https://pith.science/pith/CCEILMG7APRY7PV5ESNLG2WN4I","download_json":"https://pith.science/pith/CCEILMG7APRY7PV5ESNLG2WN4I.json","view_paper":"https://pith.science/paper/CCEILMG7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.08360&json=true","fetch_graph":"https://pith.science/api/pith-number/CCEILMG7APRY7PV5ESNLG2WN4I/graph.json","fetch_events":"https://pith.science/api/pith-number/CCEILMG7APRY7PV5ESNLG2WN4I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CCEILMG7APRY7PV5ESNLG2WN4I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CCEILMG7APRY7PV5ESNLG2WN4I/action/storage_attestation","attest_author":"https://pith.science/pith/CCEILMG7APRY7PV5ESNLG2WN4I/action/author_attestation","sign_citation":"https://pith.science/pith/CCEILMG7APRY7PV5ESNLG2WN4I/action/citation_signature","submit_replication":"https://pith.science/pith/CCEILMG7APRY7PV5ESNLG2WN4I/action/replication_record"}},"created_at":"2026-07-05T11:18:48.502640+00:00","updated_at":"2026-07-05T11:18:48.502640+00:00"}