{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:MBNTRUP47YKM4NDRBGOSUJ3U64","short_pith_number":"pith:MBNTRUP4","schema_version":"1.0","canonical_sha256":"605b38d1fcfe14ce3471099d2a2774f73170e5674477e044ce046762fdc2a4bc","source":{"kind":"arxiv","id":"1809.08587","version":4},"attestation_state":"computed","paper":{"title":"Exponential Convergence Time of Gradient Descent for One-Dimensional Deep Linear Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ohad Shamir","submitted_at":"2018-09-23T12:32:45Z","abstract_excerpt":"We study the dynamics of gradient descent on objective functions of the form $f(\\prod_{i=1}^{k} w_i)$ (with respect to scalar parameters $w_1,\\ldots,w_k$), which arise in the context of training depth-$k$ linear neural networks. We prove that for standard random initializations, and under mild assumptions on $f$, the number of iterations required for convergence scales exponentially with the depth $k$. We also show empirically that this phenomenon can occur in higher dimensions, where each $w_i$ is a matrix. This highlights a potential obstacle in understanding the convergence of gradient-base"},"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":"1809.08587","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-09-23T12:32:45Z","cross_cats_sorted":["cs.NE","math.OC","stat.ML"],"title_canon_sha256":"4cea27c5e6772158e1df8f56770a6b94f6767822742538d045e9a4049e21783d","abstract_canon_sha256":"7682a179b6c92b15f437f65dc528628bc78f9e07d0453c9d0c01c6bf609f3fcb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:43:28.244040Z","signature_b64":"ZGOOOv+UUrCxQrVBVdgoOurTfH0ogCNXLp/iavlsg5nY0ERJwh1TZUkf9DT8VuiZgN2nw1ZFODIrKHRjOQ6dCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"605b38d1fcfe14ce3471099d2a2774f73170e5674477e044ce046762fdc2a4bc","last_reissued_at":"2026-05-17T23:43:28.243609Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:43:28.243609Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Exponential Convergence Time of Gradient Descent for One-Dimensional Deep Linear Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ohad Shamir","submitted_at":"2018-09-23T12:32:45Z","abstract_excerpt":"We study the dynamics of gradient descent on objective functions of the form $f(\\prod_{i=1}^{k} w_i)$ (with respect to scalar parameters $w_1,\\ldots,w_k$), which arise in the context of training depth-$k$ linear neural networks. We prove that for standard random initializations, and under mild assumptions on $f$, the number of iterations required for convergence scales exponentially with the depth $k$. We also show empirically that this phenomenon can occur in higher dimensions, where each $w_i$ is a matrix. This highlights a potential obstacle in understanding the convergence of gradient-base"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1809.08587","kind":"arxiv","version":4},"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":"1809.08587","created_at":"2026-05-17T23:43:28.243675+00:00"},{"alias_kind":"arxiv_version","alias_value":"1809.08587v4","created_at":"2026-05-17T23:43:28.243675+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1809.08587","created_at":"2026-05-17T23:43:28.243675+00:00"},{"alias_kind":"pith_short_12","alias_value":"MBNTRUP47YKM","created_at":"2026-05-18T12:32:37.024351+00:00"},{"alias_kind":"pith_short_16","alias_value":"MBNTRUP47YKM4NDR","created_at":"2026-05-18T12:32:37.024351+00:00"},{"alias_kind":"pith_short_8","alias_value":"MBNTRUP4","created_at":"2026-05-18T12:32:37.024351+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1907.10732","citing_title":"Hessian based analysis of SGD for Deep Nets: Dynamics and Generalization","ref_index":67,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MBNTRUP47YKM4NDRBGOSUJ3U64","json":"https://pith.science/pith/MBNTRUP47YKM4NDRBGOSUJ3U64.json","graph_json":"https://pith.science/api/pith-number/MBNTRUP47YKM4NDRBGOSUJ3U64/graph.json","events_json":"https://pith.science/api/pith-number/MBNTRUP47YKM4NDRBGOSUJ3U64/events.json","paper":"https://pith.science/paper/MBNTRUP4"},"agent_actions":{"view_html":"https://pith.science/pith/MBNTRUP47YKM4NDRBGOSUJ3U64","download_json":"https://pith.science/pith/MBNTRUP47YKM4NDRBGOSUJ3U64.json","view_paper":"https://pith.science/paper/MBNTRUP4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1809.08587&json=true","fetch_graph":"https://pith.science/api/pith-number/MBNTRUP47YKM4NDRBGOSUJ3U64/graph.json","fetch_events":"https://pith.science/api/pith-number/MBNTRUP47YKM4NDRBGOSUJ3U64/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MBNTRUP47YKM4NDRBGOSUJ3U64/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MBNTRUP47YKM4NDRBGOSUJ3U64/action/storage_attestation","attest_author":"https://pith.science/pith/MBNTRUP47YKM4NDRBGOSUJ3U64/action/author_attestation","sign_citation":"https://pith.science/pith/MBNTRUP47YKM4NDRBGOSUJ3U64/action/citation_signature","submit_replication":"https://pith.science/pith/MBNTRUP47YKM4NDRBGOSUJ3U64/action/replication_record"}},"created_at":"2026-05-17T23:43:28.243675+00:00","updated_at":"2026-05-17T23:43:28.243675+00:00"}