{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:6K6S3IEQ557RDEK2Y4DQGPOJIS","short_pith_number":"pith:6K6S3IEQ","schema_version":"1.0","canonical_sha256":"f2bd2da090ef7f11915ac707033dc94484714757a27b5ea03d182d82f805f9d8","source":{"kind":"arxiv","id":"2002.03305","version":2},"attestation_state":"computed","paper":{"title":"Momentum Improves Normalized SGD","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ashok Cutkosky, Harsh Mehta","submitted_at":"2020-02-09T07:00:54Z","abstract_excerpt":"We provide an improved analysis of normalized SGD showing that adding momentum provably removes the need for large batch sizes on non-convex objectives. Then, we consider the case of objectives with bounded second derivative and show that in this case a small tweak to the momentum formula allows normalized SGD with momentum to find an $\\epsilon$-critical point in $O(1/\\epsilon^{3.5})$ iterations, matching the best-known rates without accruing any logarithmic factors or dependence on dimension. We also provide an adaptive method that automatically improves convergence rates when the variance in"},"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":"2002.03305","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-09T07:00:54Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"0daf4decb73da41248b5a32ad77dfa8f4e349017a787588910c0f5f6a95a70fd","abstract_canon_sha256":"5e0d7d01523839ee90dc4628be58bfb3a5b7985cac970e1ee43c9d234dd86e8c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:03:23.228349Z","signature_b64":"SzPrw8lsMCwk1aMF3o/kotZmtEk3vUFvTLT6ROcjolb+ygyfzj7HEHZ434eaRHczxcDjP6cimvKWW5R6vl/kBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f2bd2da090ef7f11915ac707033dc94484714757a27b5ea03d182d82f805f9d8","last_reissued_at":"2026-07-05T01:03:23.227875Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:03:23.227875Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Momentum Improves Normalized SGD","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ashok Cutkosky, Harsh Mehta","submitted_at":"2020-02-09T07:00:54Z","abstract_excerpt":"We provide an improved analysis of normalized SGD showing that adding momentum provably removes the need for large batch sizes on non-convex objectives. Then, we consider the case of objectives with bounded second derivative and show that in this case a small tweak to the momentum formula allows normalized SGD with momentum to find an $\\epsilon$-critical point in $O(1/\\epsilon^{3.5})$ iterations, matching the best-known rates without accruing any logarithmic factors or dependence on dimension. We also provide an adaptive method that automatically improves convergence rates when the variance in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.03305","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/2002.03305/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":"2002.03305","created_at":"2026-07-05T01:03:23.227931+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.03305v2","created_at":"2026-07-05T01:03:23.227931+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.03305","created_at":"2026-07-05T01:03:23.227931+00:00"},{"alias_kind":"pith_short_12","alias_value":"6K6S3IEQ557R","created_at":"2026-07-05T01:03:23.227931+00:00"},{"alias_kind":"pith_short_16","alias_value":"6K6S3IEQ557RDEK2","created_at":"2026-07-05T01:03:23.227931+00:00"},{"alias_kind":"pith_short_8","alias_value":"6K6S3IEQ","created_at":"2026-07-05T01:03:23.227931+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/6K6S3IEQ557RDEK2Y4DQGPOJIS","json":"https://pith.science/pith/6K6S3IEQ557RDEK2Y4DQGPOJIS.json","graph_json":"https://pith.science/api/pith-number/6K6S3IEQ557RDEK2Y4DQGPOJIS/graph.json","events_json":"https://pith.science/api/pith-number/6K6S3IEQ557RDEK2Y4DQGPOJIS/events.json","paper":"https://pith.science/paper/6K6S3IEQ"},"agent_actions":{"view_html":"https://pith.science/pith/6K6S3IEQ557RDEK2Y4DQGPOJIS","download_json":"https://pith.science/pith/6K6S3IEQ557RDEK2Y4DQGPOJIS.json","view_paper":"https://pith.science/paper/6K6S3IEQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.03305&json=true","fetch_graph":"https://pith.science/api/pith-number/6K6S3IEQ557RDEK2Y4DQGPOJIS/graph.json","fetch_events":"https://pith.science/api/pith-number/6K6S3IEQ557RDEK2Y4DQGPOJIS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6K6S3IEQ557RDEK2Y4DQGPOJIS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6K6S3IEQ557RDEK2Y4DQGPOJIS/action/storage_attestation","attest_author":"https://pith.science/pith/6K6S3IEQ557RDEK2Y4DQGPOJIS/action/author_attestation","sign_citation":"https://pith.science/pith/6K6S3IEQ557RDEK2Y4DQGPOJIS/action/citation_signature","submit_replication":"https://pith.science/pith/6K6S3IEQ557RDEK2Y4DQGPOJIS/action/replication_record"}},"created_at":"2026-07-05T01:03:23.227931+00:00","updated_at":"2026-07-05T01:03:23.227931+00:00"}