{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QRP64NFLVGLCICBM2KLFSZNNKS","short_pith_number":"pith:QRP64NFL","schema_version":"1.0","canonical_sha256":"845fee34aba99624082cd2965965ad54946de055cd7ec013686a8a556b04e136","source":{"kind":"arxiv","id":"2502.07488","version":1},"attestation_state":"computed","paper":{"title":"Improving Adaptive Moment Optimization via Preconditioner Diagonalization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Liu, Lizhang Chen, Qiang Liu, Son Nguyen","submitted_at":"2025-02-11T11:48:04Z","abstract_excerpt":"Modern adaptive optimization methods, such as Adam and its variants, have emerged as the most widely used tools in deep learning over recent years. These algorithms offer automatic mechanisms for dynamically adjusting the update step based on estimates of gradient statistics. Compared to traditional algorithms like Stochastic Gradient Descent, these adaptive methods are typically more robust to model scale and hyperparameter tuning. However, the gradient statistics employed by these methods often do not leverage sufficient gradient covariance information, leading to suboptimal updates in certa"},"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":"2502.07488","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-11T11:48:04Z","cross_cats_sorted":[],"title_canon_sha256":"2a2489df08533b1d786a5ce886f749c727cb1b8e79c83f1e1e1b54efe60e6cb4","abstract_canon_sha256":"d1ad87505428840542111c091631330df42ad71e19af6f2284a5060be18c14b2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:12:38.223579Z","signature_b64":"TAmE5U7v/8LenLpMlfHajhOONHzIfFKxKBAbxvfnOfY9R+iw5//HeaPfvhiwzEPciyAuJM8bq6TbWrlxuizkBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"845fee34aba99624082cd2965965ad54946de055cd7ec013686a8a556b04e136","last_reissued_at":"2026-07-05T10:12:38.223083Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:12:38.223083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Adaptive Moment Optimization via Preconditioner Diagonalization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Liu, Lizhang Chen, Qiang Liu, Son Nguyen","submitted_at":"2025-02-11T11:48:04Z","abstract_excerpt":"Modern adaptive optimization methods, such as Adam and its variants, have emerged as the most widely used tools in deep learning over recent years. These algorithms offer automatic mechanisms for dynamically adjusting the update step based on estimates of gradient statistics. Compared to traditional algorithms like Stochastic Gradient Descent, these adaptive methods are typically more robust to model scale and hyperparameter tuning. However, the gradient statistics employed by these methods often do not leverage sufficient gradient covariance information, leading to suboptimal updates in certa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.07488","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/2502.07488/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":"2502.07488","created_at":"2026-07-05T10:12:38.223156+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.07488v1","created_at":"2026-07-05T10:12:38.223156+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.07488","created_at":"2026-07-05T10:12:38.223156+00:00"},{"alias_kind":"pith_short_12","alias_value":"QRP64NFLVGLC","created_at":"2026-07-05T10:12:38.223156+00:00"},{"alias_kind":"pith_short_16","alias_value":"QRP64NFLVGLCICBM","created_at":"2026-07-05T10:12:38.223156+00:00"},{"alias_kind":"pith_short_8","alias_value":"QRP64NFL","created_at":"2026-07-05T10:12:38.223156+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.12968","citing_title":"Evolution of Optimization Methods: Algorithms, Scenarios, and Evaluations","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QRP64NFLVGLCICBM2KLFSZNNKS","json":"https://pith.science/pith/QRP64NFLVGLCICBM2KLFSZNNKS.json","graph_json":"https://pith.science/api/pith-number/QRP64NFLVGLCICBM2KLFSZNNKS/graph.json","events_json":"https://pith.science/api/pith-number/QRP64NFLVGLCICBM2KLFSZNNKS/events.json","paper":"https://pith.science/paper/QRP64NFL"},"agent_actions":{"view_html":"https://pith.science/pith/QRP64NFLVGLCICBM2KLFSZNNKS","download_json":"https://pith.science/pith/QRP64NFLVGLCICBM2KLFSZNNKS.json","view_paper":"https://pith.science/paper/QRP64NFL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.07488&json=true","fetch_graph":"https://pith.science/api/pith-number/QRP64NFLVGLCICBM2KLFSZNNKS/graph.json","fetch_events":"https://pith.science/api/pith-number/QRP64NFLVGLCICBM2KLFSZNNKS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QRP64NFLVGLCICBM2KLFSZNNKS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QRP64NFLVGLCICBM2KLFSZNNKS/action/storage_attestation","attest_author":"https://pith.science/pith/QRP64NFLVGLCICBM2KLFSZNNKS/action/author_attestation","sign_citation":"https://pith.science/pith/QRP64NFLVGLCICBM2KLFSZNNKS/action/citation_signature","submit_replication":"https://pith.science/pith/QRP64NFLVGLCICBM2KLFSZNNKS/action/replication_record"}},"created_at":"2026-07-05T10:12:38.223156+00:00","updated_at":"2026-07-05T10:12:38.223156+00:00"}