{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:ORPICM32QOFK5GOUJZ47DF472P","short_pith_number":"pith:ORPICM32","schema_version":"1.0","canonical_sha256":"745e81337a838aae99d44e79f1979fd3c7106af57557ba571639e12903d18d32","source":{"kind":"arxiv","id":"1804.00325","version":3},"attestation_state":"computed","paper":{"title":"Aggregated Momentum: Stability Through Passive Damping","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"James Lucas, Richard Zemel, Roger Grosse, Shengyang Sun","submitted_at":"2018-04-01T17:53:03Z","abstract_excerpt":"Momentum is a simple and widely used trick which allows gradient-based optimizers to pick up speed along low curvature directions. Its performance depends crucially on a damping coefficient $\\beta$. Large $\\beta$ values can potentially deliver much larger speedups, but are prone to oscillations and instability; hence one typically resorts to small values such as 0.5 or 0.9. We propose Aggregated Momentum (AggMo), a variant of momentum which combines multiple velocity vectors with different $\\beta$ parameters. AggMo is trivial to implement, but significantly dampens oscillations, enabling it to"},"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":"1804.00325","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-04-01T17:53:03Z","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"title_canon_sha256":"9b1a80e49b8732efbb26f36c6ef71da946fcaf66658c77c117352738dc1eebce","abstract_canon_sha256":"052020463dfb6163c60e604ee652133fc3c91f08a42df66a17e89db5805dbe99"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:47:16.057825Z","signature_b64":"DTm6C7X8GZm2MK04NnbO47atyuec3xqsfOIuP4rcmDn0kVI+Awv7hdz9r8mAj78dsTxbU5E4RUqgBmO2zrs7DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"745e81337a838aae99d44e79f1979fd3c7106af57557ba571639e12903d18d32","last_reissued_at":"2026-05-17T23:47:16.057332Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:47:16.057332Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Aggregated Momentum: Stability Through Passive Damping","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"James Lucas, Richard Zemel, Roger Grosse, Shengyang Sun","submitted_at":"2018-04-01T17:53:03Z","abstract_excerpt":"Momentum is a simple and widely used trick which allows gradient-based optimizers to pick up speed along low curvature directions. Its performance depends crucially on a damping coefficient $\\beta$. Large $\\beta$ values can potentially deliver much larger speedups, but are prone to oscillations and instability; hence one typically resorts to small values such as 0.5 or 0.9. We propose Aggregated Momentum (AggMo), a variant of momentum which combines multiple velocity vectors with different $\\beta$ parameters. AggMo is trivial to implement, but significantly dampens oscillations, enabling it to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1804.00325","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"1804.00325","created_at":"2026-05-17T23:47:16.057405+00:00"},{"alias_kind":"arxiv_version","alias_value":"1804.00325v3","created_at":"2026-05-17T23:47:16.057405+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1804.00325","created_at":"2026-05-17T23:47:16.057405+00:00"},{"alias_kind":"pith_short_12","alias_value":"ORPICM32QOFK","created_at":"2026-05-18T12:32:43.782077+00:00"},{"alias_kind":"pith_short_16","alias_value":"ORPICM32QOFK5GOU","created_at":"2026-05-18T12:32:43.782077+00:00"},{"alias_kind":"pith_short_8","alias_value":"ORPICM32","created_at":"2026-05-18T12:32:43.782077+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2307.07030","citing_title":"Accelerated Gradient Methods for Nonconvex Optimization: Escape Trajectories From Strict Saddle Points and Convergence to Local Minima","ref_index":54,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ORPICM32QOFK5GOUJZ47DF472P","json":"https://pith.science/pith/ORPICM32QOFK5GOUJZ47DF472P.json","graph_json":"https://pith.science/api/pith-number/ORPICM32QOFK5GOUJZ47DF472P/graph.json","events_json":"https://pith.science/api/pith-number/ORPICM32QOFK5GOUJZ47DF472P/events.json","paper":"https://pith.science/paper/ORPICM32"},"agent_actions":{"view_html":"https://pith.science/pith/ORPICM32QOFK5GOUJZ47DF472P","download_json":"https://pith.science/pith/ORPICM32QOFK5GOUJZ47DF472P.json","view_paper":"https://pith.science/paper/ORPICM32","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1804.00325&json=true","fetch_graph":"https://pith.science/api/pith-number/ORPICM32QOFK5GOUJZ47DF472P/graph.json","fetch_events":"https://pith.science/api/pith-number/ORPICM32QOFK5GOUJZ47DF472P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ORPICM32QOFK5GOUJZ47DF472P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ORPICM32QOFK5GOUJZ47DF472P/action/storage_attestation","attest_author":"https://pith.science/pith/ORPICM32QOFK5GOUJZ47DF472P/action/author_attestation","sign_citation":"https://pith.science/pith/ORPICM32QOFK5GOUJZ47DF472P/action/citation_signature","submit_replication":"https://pith.science/pith/ORPICM32QOFK5GOUJZ47DF472P/action/replication_record"}},"created_at":"2026-05-17T23:47:16.057405+00:00","updated_at":"2026-05-17T23:47:16.057405+00:00"}