{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:WMIJJVJVB4F7742PJF7LG33NIU","short_pith_number":"pith:WMIJJVJV","canonical_record":{"source":{"id":"2110.03677","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-07T17:58:21Z","cross_cats_sorted":["math.DS","math.OC"],"title_canon_sha256":"94240fed2f816ef7cc7e4638efe72379c84a042b076bfde046a00534b500f733","abstract_canon_sha256":"d8201836318d8993682b245b8ff2c01b1d207127f8306385de8014b5394825be"},"schema_version":"1.0"},"canonical_sha256":"b31094d5350f0bfff34f497eb36f6d452569b92507443821e45cd6a813c6169b","source":{"kind":"arxiv","id":"2110.03677","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2110.03677","created_at":"2026-07-05T04:00:18Z"},{"alias_kind":"arxiv_version","alias_value":"2110.03677v2","created_at":"2026-07-05T04:00:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.03677","created_at":"2026-07-05T04:00:18Z"},{"alias_kind":"pith_short_12","alias_value":"WMIJJVJVB4F7","created_at":"2026-07-05T04:00:18Z"},{"alias_kind":"pith_short_16","alias_value":"WMIJJVJVB4F7742P","created_at":"2026-07-05T04:00:18Z"},{"alias_kind":"pith_short_8","alias_value":"WMIJJVJV","created_at":"2026-07-05T04:00:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:WMIJJVJVB4F7742PJF7LG33NIU","target":"record","payload":{"canonical_record":{"source":{"id":"2110.03677","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-07T17:58:21Z","cross_cats_sorted":["math.DS","math.OC"],"title_canon_sha256":"94240fed2f816ef7cc7e4638efe72379c84a042b076bfde046a00534b500f733","abstract_canon_sha256":"d8201836318d8993682b245b8ff2c01b1d207127f8306385de8014b5394825be"},"schema_version":"1.0"},"canonical_sha256":"b31094d5350f0bfff34f497eb36f6d452569b92507443821e45cd6a813c6169b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:00:18.703717Z","signature_b64":"hUOfP/4JSUXdxjgBzeogOxxf73w/DqKYjK9wd5GttiRGNP440kBkHxe4A/akqXYrybKIhFkOlFmkL2D8p/VcDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b31094d5350f0bfff34f497eb36f6d452569b92507443821e45cd6a813c6169b","last_reissued_at":"2026-07-05T04:00:18.703248Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:00:18.703248Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2110.03677","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:00:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3n+WUTnWvMoGwPxkiS5SBZOnJdvP6YsWmR41Uzzh+6JOZ3MbTB9/zBo9j5hwaVSUb+YZ0DDx61KxtlOzwGBzAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T08:51:49.240127Z"},"content_sha256":"95cfc0186974bc942cd74a16b0926638774ca2287e9f31011e4bfba9c6e38aaa","schema_version":"1.0","event_id":"sha256:95cfc0186974bc942cd74a16b0926638774ca2287e9f31011e4bfba9c6e38aaa"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:WMIJJVJVB4F7742PJF7LG33NIU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Large Learning Rate Tames Homogeneity: Convergence and Balancing Effect","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.DS","math.OC"],"primary_cat":"cs.LG","authors_text":"Minshuo Chen, Molei Tao, Tuo Zhao, Yuqing Wang","submitted_at":"2021-10-07T17:58:21Z","abstract_excerpt":"Recent empirical advances show that training deep models with large learning rate often improves generalization performance. However, theoretical justifications on the benefits of large learning rate are highly limited, due to challenges in analysis. In this paper, we consider using Gradient Descent (GD) with a large learning rate on a homogeneous matrix factorization problem, i.e., $\\min_{X, Y} \\|A - XY^\\top\\|_{\\sf F}^2$. We prove a convergence theory for constant large learning rates well beyond $2/L$, where $L$ is the largest eigenvalue of Hessian at the initialization. Moreover, we rigorou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.03677","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/2110.03677/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:00:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SPQLyN3zG22m4OFCkFi+8JbPhvpLaB+JGatInarrXXuIGcSq9lE4dmprkiOwIIQ/y9Cko+BZUgeoPDfVYWEmBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T08:51:49.240658Z"},"content_sha256":"1d4c1e4370aaeb65b712dc5ee2f8e1967288796ddf5357222363ca3de7003454","schema_version":"1.0","event_id":"sha256:1d4c1e4370aaeb65b712dc5ee2f8e1967288796ddf5357222363ca3de7003454"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WMIJJVJVB4F7742PJF7LG33NIU/bundle.json","state_url":"https://pith.science/pith/WMIJJVJVB4F7742PJF7LG33NIU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WMIJJVJVB4F7742PJF7LG33NIU/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-18T08:51:49Z","links":{"resolver":"https://pith.science/pith/WMIJJVJVB4F7742PJF7LG33NIU","bundle":"https://pith.science/pith/WMIJJVJVB4F7742PJF7LG33NIU/bundle.json","state":"https://pith.science/pith/WMIJJVJVB4F7742PJF7LG33NIU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WMIJJVJVB4F7742PJF7LG33NIU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:WMIJJVJVB4F7742PJF7LG33NIU","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"d8201836318d8993682b245b8ff2c01b1d207127f8306385de8014b5394825be","cross_cats_sorted":["math.DS","math.OC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-07T17:58:21Z","title_canon_sha256":"94240fed2f816ef7cc7e4638efe72379c84a042b076bfde046a00534b500f733"},"schema_version":"1.0","source":{"id":"2110.03677","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2110.03677","created_at":"2026-07-05T04:00:18Z"},{"alias_kind":"arxiv_version","alias_value":"2110.03677v2","created_at":"2026-07-05T04:00:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.03677","created_at":"2026-07-05T04:00:18Z"},{"alias_kind":"pith_short_12","alias_value":"WMIJJVJVB4F7","created_at":"2026-07-05T04:00:18Z"},{"alias_kind":"pith_short_16","alias_value":"WMIJJVJVB4F7742P","created_at":"2026-07-05T04:00:18Z"},{"alias_kind":"pith_short_8","alias_value":"WMIJJVJV","created_at":"2026-07-05T04:00:18Z"}],"graph_snapshots":[{"event_id":"sha256:1d4c1e4370aaeb65b712dc5ee2f8e1967288796ddf5357222363ca3de7003454","target":"graph","created_at":"2026-07-05T04:00:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2110.03677/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent empirical advances show that training deep models with large learning rate often improves generalization performance. However, theoretical justifications on the benefits of large learning rate are highly limited, due to challenges in analysis. In this paper, we consider using Gradient Descent (GD) with a large learning rate on a homogeneous matrix factorization problem, i.e., $\\min_{X, Y} \\|A - XY^\\top\\|_{\\sf F}^2$. We prove a convergence theory for constant large learning rates well beyond $2/L$, where $L$ is the largest eigenvalue of Hessian at the initialization. Moreover, we rigorou","authors_text":"Minshuo Chen, Molei Tao, Tuo Zhao, Yuqing Wang","cross_cats":["math.DS","math.OC"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-07T17:58:21Z","title":"Large Learning Rate Tames Homogeneity: Convergence and Balancing Effect"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.03677","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:95cfc0186974bc942cd74a16b0926638774ca2287e9f31011e4bfba9c6e38aaa","target":"record","created_at":"2026-07-05T04:00:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"d8201836318d8993682b245b8ff2c01b1d207127f8306385de8014b5394825be","cross_cats_sorted":["math.DS","math.OC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-07T17:58:21Z","title_canon_sha256":"94240fed2f816ef7cc7e4638efe72379c84a042b076bfde046a00534b500f733"},"schema_version":"1.0","source":{"id":"2110.03677","kind":"arxiv","version":2}},"canonical_sha256":"b31094d5350f0bfff34f497eb36f6d452569b92507443821e45cd6a813c6169b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b31094d5350f0bfff34f497eb36f6d452569b92507443821e45cd6a813c6169b","first_computed_at":"2026-07-05T04:00:18.703248Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:00:18.703248Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hUOfP/4JSUXdxjgBzeogOxxf73w/DqKYjK9wd5GttiRGNP440kBkHxe4A/akqXYrybKIhFkOlFmkL2D8p/VcDg==","signature_status":"signed_v1","signed_at":"2026-07-05T04:00:18.703717Z","signed_message":"canonical_sha256_bytes"},"source_id":"2110.03677","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:95cfc0186974bc942cd74a16b0926638774ca2287e9f31011e4bfba9c6e38aaa","sha256:1d4c1e4370aaeb65b712dc5ee2f8e1967288796ddf5357222363ca3de7003454"],"state_sha256":"531151a1656e0d66d529d7ce450693e37db3af8b316f1a6968573a912bdba7e0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DS8X0wltYWibg0UgfhwuTflsbYQlmDFf5DWr73Y0BwRM6ioponHYyQ88C7H9DAOZkSh4TuaOoyIAxHleJ811Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T08:51:49.246488Z","bundle_sha256":"7919e7d84b8816fed415eeea93770fe84f1926271a9f861cd3e324f05a38b605"}}