{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:UKF2QKIQ7ZJV6G3OVV6LKFYFOE","short_pith_number":"pith:UKF2QKIQ","schema_version":"1.0","canonical_sha256":"a28ba82910fe535f1b6ead7cb51705712a6a538030c1ba762ea40930fc8ffd75","source":{"kind":"arxiv","id":"2403.02648","version":4},"attestation_state":"computed","paper":{"title":"Remove that Square Root: A New Efficient Scale-Invariant Version of AdaGrad","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC"],"primary_cat":"cs.LG","authors_text":"Eduard Gorbunov, Martin Takac, Nazarii Tupitsa, Nicolas Loizou, Samuel Horvath, Sayantan Choudhury","submitted_at":"2024-03-05T04:35:59Z","abstract_excerpt":"Adaptive methods are extremely popular in machine learning as they make learning rate tuning less expensive. This paper introduces a novel optimization algorithm named KATE, which presents a scale-invariant adaptation of the well-known AdaGrad algorithm. We prove the scale-invariance of KATE for the case of Generalized Linear Models. Moreover, for general smooth non-convex problems, we establish a convergence rate of $O \\left(\\frac{\\log T}{\\sqrt{T}} \\right)$ for KATE, matching the best-known ones for AdaGrad and Adam. We also compare KATE to other state-of-the-art adaptive algorithms Adam and "},"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":"2403.02648","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-05T04:35:59Z","cross_cats_sorted":["cs.AI","math.OC"],"title_canon_sha256":"9fef324c0442baaccfbb60f251edc5cba99912dbe4c1df774740131c5d879e3a","abstract_canon_sha256":"07c14f5624b5b016fffa5e0c504556bca0eb92bf06b44604aa12c13322152ac8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:00:35.106834Z","signature_b64":"3OULNPljKAksrFKoLheTMncaSbfNlDJrqRasClaXgAxRPgciEaLAiaUNkpfPpIzYwQWh8wyevkQ2f8jdkCa3Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a28ba82910fe535f1b6ead7cb51705712a6a538030c1ba762ea40930fc8ffd75","last_reissued_at":"2026-07-05T10:00:35.106494Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:00:35.106494Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Remove that Square Root: A New Efficient Scale-Invariant Version of AdaGrad","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC"],"primary_cat":"cs.LG","authors_text":"Eduard Gorbunov, Martin Takac, Nazarii Tupitsa, Nicolas Loizou, Samuel Horvath, Sayantan Choudhury","submitted_at":"2024-03-05T04:35:59Z","abstract_excerpt":"Adaptive methods are extremely popular in machine learning as they make learning rate tuning less expensive. This paper introduces a novel optimization algorithm named KATE, which presents a scale-invariant adaptation of the well-known AdaGrad algorithm. We prove the scale-invariance of KATE for the case of Generalized Linear Models. Moreover, for general smooth non-convex problems, we establish a convergence rate of $O \\left(\\frac{\\log T}{\\sqrt{T}} \\right)$ for KATE, matching the best-known ones for AdaGrad and Adam. We also compare KATE to other state-of-the-art adaptive algorithms Adam and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.02648","kind":"arxiv","version":4},"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/2403.02648/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":"2403.02648","created_at":"2026-07-05T10:00:35.106548+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.02648v4","created_at":"2026-07-05T10:00:35.106548+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.02648","created_at":"2026-07-05T10:00:35.106548+00:00"},{"alias_kind":"pith_short_12","alias_value":"UKF2QKIQ7ZJV","created_at":"2026-07-05T10:00:35.106548+00:00"},{"alias_kind":"pith_short_16","alias_value":"UKF2QKIQ7ZJV6G3O","created_at":"2026-07-05T10:00:35.106548+00:00"},{"alias_kind":"pith_short_8","alias_value":"UKF2QKIQ","created_at":"2026-07-05T10:00:35.106548+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.17288","citing_title":"Polyak Stepsize: Estimating Optimal Functional Values Without Parameters or Prior Knowledge","ref_index":2011,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UKF2QKIQ7ZJV6G3OVV6LKFYFOE","json":"https://pith.science/pith/UKF2QKIQ7ZJV6G3OVV6LKFYFOE.json","graph_json":"https://pith.science/api/pith-number/UKF2QKIQ7ZJV6G3OVV6LKFYFOE/graph.json","events_json":"https://pith.science/api/pith-number/UKF2QKIQ7ZJV6G3OVV6LKFYFOE/events.json","paper":"https://pith.science/paper/UKF2QKIQ"},"agent_actions":{"view_html":"https://pith.science/pith/UKF2QKIQ7ZJV6G3OVV6LKFYFOE","download_json":"https://pith.science/pith/UKF2QKIQ7ZJV6G3OVV6LKFYFOE.json","view_paper":"https://pith.science/paper/UKF2QKIQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.02648&json=true","fetch_graph":"https://pith.science/api/pith-number/UKF2QKIQ7ZJV6G3OVV6LKFYFOE/graph.json","fetch_events":"https://pith.science/api/pith-number/UKF2QKIQ7ZJV6G3OVV6LKFYFOE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UKF2QKIQ7ZJV6G3OVV6LKFYFOE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UKF2QKIQ7ZJV6G3OVV6LKFYFOE/action/storage_attestation","attest_author":"https://pith.science/pith/UKF2QKIQ7ZJV6G3OVV6LKFYFOE/action/author_attestation","sign_citation":"https://pith.science/pith/UKF2QKIQ7ZJV6G3OVV6LKFYFOE/action/citation_signature","submit_replication":"https://pith.science/pith/UKF2QKIQ7ZJV6G3OVV6LKFYFOE/action/replication_record"}},"created_at":"2026-07-05T10:00:35.106548+00:00","updated_at":"2026-07-05T10:00:35.106548+00:00"}