{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:I2IXS2OP3W45ZB2E744RAI3KGG","short_pith_number":"pith:I2IXS2OP","schema_version":"1.0","canonical_sha256":"46917969cfddb9dc8744ff3910236a31bf315b480b81ba9e34c20605891ea496","source":{"kind":"arxiv","id":"2107.00595","version":2},"attestation_state":"computed","paper":{"title":"Fast Margin Maximization via Dual Acceleration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Matus Telgarsky, Nathan Srebro, Ziwei Ji","submitted_at":"2021-07-01T16:36:39Z","abstract_excerpt":"We present and analyze a momentum-based gradient method for training linear classifiers with an exponentially-tailed loss (e.g., the exponential or logistic loss), which maximizes the classification margin on separable data at a rate of $\\widetilde{\\mathcal{O}}(1/t^2)$. This contrasts with a rate of $\\mathcal{O}(1/\\log(t))$ for standard gradient descent, and $\\mathcal{O}(1/t)$ for normalized gradient descent. This momentum-based method is derived via the convex dual of the maximum-margin problem, and specifically by applying Nesterov acceleration to this dual, which manages to result in a simp"},"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":"2107.00595","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-01T16:36:39Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"741c9f46789a65950609b1eb5cde3b3595f1db7e51cef7e2198830ef3928ab1f","abstract_canon_sha256":"7f2bc73cfd3b66485b053aa32928cf8412ce6a8be7b93edb064f328d90b0d7c9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:07:50.456361Z","signature_b64":"icMOhPoOAJhWaiC4aduHL3EMRtLptpalL5co8ooZzjYsgQ7FZtll0OX0K0MZ3W7fTuTpqLpRt56c4N7L7Z9iDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"46917969cfddb9dc8744ff3910236a31bf315b480b81ba9e34c20605891ea496","last_reissued_at":"2026-07-05T03:07:50.455928Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:07:50.455928Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fast Margin Maximization via Dual Acceleration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Matus Telgarsky, Nathan Srebro, Ziwei Ji","submitted_at":"2021-07-01T16:36:39Z","abstract_excerpt":"We present and analyze a momentum-based gradient method for training linear classifiers with an exponentially-tailed loss (e.g., the exponential or logistic loss), which maximizes the classification margin on separable data at a rate of $\\widetilde{\\mathcal{O}}(1/t^2)$. This contrasts with a rate of $\\mathcal{O}(1/\\log(t))$ for standard gradient descent, and $\\mathcal{O}(1/t)$ for normalized gradient descent. This momentum-based method is derived via the convex dual of the maximum-margin problem, and specifically by applying Nesterov acceleration to this dual, which manages to result in a simp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.00595","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/2107.00595/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":"2107.00595","created_at":"2026-07-05T03:07:50.455991+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.00595v2","created_at":"2026-07-05T03:07:50.455991+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.00595","created_at":"2026-07-05T03:07:50.455991+00:00"},{"alias_kind":"pith_short_12","alias_value":"I2IXS2OP3W45","created_at":"2026-07-05T03:07:50.455991+00:00"},{"alias_kind":"pith_short_16","alias_value":"I2IXS2OP3W45ZB2E","created_at":"2026-07-05T03:07:50.455991+00:00"},{"alias_kind":"pith_short_8","alias_value":"I2IXS2OP","created_at":"2026-07-05T03:07:50.455991+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/I2IXS2OP3W45ZB2E744RAI3KGG","json":"https://pith.science/pith/I2IXS2OP3W45ZB2E744RAI3KGG.json","graph_json":"https://pith.science/api/pith-number/I2IXS2OP3W45ZB2E744RAI3KGG/graph.json","events_json":"https://pith.science/api/pith-number/I2IXS2OP3W45ZB2E744RAI3KGG/events.json","paper":"https://pith.science/paper/I2IXS2OP"},"agent_actions":{"view_html":"https://pith.science/pith/I2IXS2OP3W45ZB2E744RAI3KGG","download_json":"https://pith.science/pith/I2IXS2OP3W45ZB2E744RAI3KGG.json","view_paper":"https://pith.science/paper/I2IXS2OP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.00595&json=true","fetch_graph":"https://pith.science/api/pith-number/I2IXS2OP3W45ZB2E744RAI3KGG/graph.json","fetch_events":"https://pith.science/api/pith-number/I2IXS2OP3W45ZB2E744RAI3KGG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I2IXS2OP3W45ZB2E744RAI3KGG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I2IXS2OP3W45ZB2E744RAI3KGG/action/storage_attestation","attest_author":"https://pith.science/pith/I2IXS2OP3W45ZB2E744RAI3KGG/action/author_attestation","sign_citation":"https://pith.science/pith/I2IXS2OP3W45ZB2E744RAI3KGG/action/citation_signature","submit_replication":"https://pith.science/pith/I2IXS2OP3W45ZB2E744RAI3KGG/action/replication_record"}},"created_at":"2026-07-05T03:07:50.455991+00:00","updated_at":"2026-07-05T03:07:50.455991+00:00"}