{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:BXOA6LOWPJZXANKBSNCSPGQ5A7","short_pith_number":"pith:BXOA6LOW","schema_version":"1.0","canonical_sha256":"0ddc0f2dd67a737035419345279a1d07ff434f50224b564a03e8a64e2fec6246","source":{"kind":"arxiv","id":"2505.16098","version":1},"attestation_state":"computed","paper":{"title":"Dimension-adapted Momentum Outscales SGD","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"stat.ML","authors_text":"Courtney Paquette, Damien Ferbach, Elliot Paquette, Gauthier Gidel, Katie Everett","submitted_at":"2025-05-22T00:58:50Z","abstract_excerpt":"We investigate scaling laws for stochastic momentum algorithms with small batch on the power law random features model, parameterized by data complexity, target complexity, and model size. When trained with a stochastic momentum algorithm, our analysis reveals four distinct loss curve shapes determined by varying data-target complexities. While traditional stochastic gradient descent with momentum (SGD-M) yields identical scaling law exponents to SGD, dimension-adapted Nesterov acceleration (DANA) improves these exponents by scaling momentum hyperparameters based on model size and data complex"},"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":"2505.16098","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2025-05-22T00:58:50Z","cross_cats_sorted":["cs.LG","math.OC"],"title_canon_sha256":"8f9d254677637b82757d0fe605fa54f9a3829ff9517f05c344a228f8c0577cb3","abstract_canon_sha256":"c3c7cc1801cbdf1156d98e673108f7efc700295c6503a578d6cfc4292bdb6c5d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:29.909038Z","signature_b64":"UOPgmGLlqXOZgwwngNXqfWgYIKoPRHVgg51OE1UjbrXGrIdwPFcvzEuLBulHnVYP+gvEu1EQZEDsLrzcAaXrAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0ddc0f2dd67a737035419345279a1d07ff434f50224b564a03e8a64e2fec6246","last_reissued_at":"2026-07-05T11:07:29.908517Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:29.908517Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dimension-adapted Momentum Outscales SGD","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"stat.ML","authors_text":"Courtney Paquette, Damien Ferbach, Elliot Paquette, Gauthier Gidel, Katie Everett","submitted_at":"2025-05-22T00:58:50Z","abstract_excerpt":"We investigate scaling laws for stochastic momentum algorithms with small batch on the power law random features model, parameterized by data complexity, target complexity, and model size. When trained with a stochastic momentum algorithm, our analysis reveals four distinct loss curve shapes determined by varying data-target complexities. While traditional stochastic gradient descent with momentum (SGD-M) yields identical scaling law exponents to SGD, dimension-adapted Nesterov acceleration (DANA) improves these exponents by scaling momentum hyperparameters based on model size and data complex"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.16098","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/2505.16098/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":"2505.16098","created_at":"2026-07-05T11:07:29.908581+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.16098v1","created_at":"2026-07-05T11:07:29.908581+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.16098","created_at":"2026-07-05T11:07:29.908581+00:00"},{"alias_kind":"pith_short_12","alias_value":"BXOA6LOWPJZX","created_at":"2026-07-05T11:07:29.908581+00:00"},{"alias_kind":"pith_short_16","alias_value":"BXOA6LOWPJZXANKB","created_at":"2026-07-05T11:07:29.908581+00:00"},{"alias_kind":"pith_short_8","alias_value":"BXOA6LOW","created_at":"2026-07-05T11:07:29.908581+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19179","citing_title":"Compute Efficiency and Serial Runtime Tradeoffs for Stochastic Momentum Methods","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2602.04774","citing_title":"Theory of Optimal Learning Rate Schedules and Scaling Laws for a Random Feature Model","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2603.26554","citing_title":"Sharp Capacity Scaling of Spectral Optimizers in Learning Associative Memory","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09552","citing_title":"Phases of Muon: When Muon Eclipses SignSGD","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BXOA6LOWPJZXANKBSNCSPGQ5A7","json":"https://pith.science/pith/BXOA6LOWPJZXANKBSNCSPGQ5A7.json","graph_json":"https://pith.science/api/pith-number/BXOA6LOWPJZXANKBSNCSPGQ5A7/graph.json","events_json":"https://pith.science/api/pith-number/BXOA6LOWPJZXANKBSNCSPGQ5A7/events.json","paper":"https://pith.science/paper/BXOA6LOW"},"agent_actions":{"view_html":"https://pith.science/pith/BXOA6LOWPJZXANKBSNCSPGQ5A7","download_json":"https://pith.science/pith/BXOA6LOWPJZXANKBSNCSPGQ5A7.json","view_paper":"https://pith.science/paper/BXOA6LOW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.16098&json=true","fetch_graph":"https://pith.science/api/pith-number/BXOA6LOWPJZXANKBSNCSPGQ5A7/graph.json","fetch_events":"https://pith.science/api/pith-number/BXOA6LOWPJZXANKBSNCSPGQ5A7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BXOA6LOWPJZXANKBSNCSPGQ5A7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BXOA6LOWPJZXANKBSNCSPGQ5A7/action/storage_attestation","attest_author":"https://pith.science/pith/BXOA6LOWPJZXANKBSNCSPGQ5A7/action/author_attestation","sign_citation":"https://pith.science/pith/BXOA6LOWPJZXANKBSNCSPGQ5A7/action/citation_signature","submit_replication":"https://pith.science/pith/BXOA6LOWPJZXANKBSNCSPGQ5A7/action/replication_record"}},"created_at":"2026-07-05T11:07:29.908581+00:00","updated_at":"2026-07-05T11:07:29.908581+00:00"}