{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:HVVEY2SHXH46YZAY5LICDKYGO2","short_pith_number":"pith:HVVEY2SH","canonical_record":{"source":{"id":"1802.06509","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-02-19T03:08:26Z","cross_cats_sorted":[],"title_canon_sha256":"cc14b4c7de347827b8bd654b047970e2e4d3a56b9487af8e289776abd165e3fc","abstract_canon_sha256":"e9a3d464cf586fc671a7a71e44bf72d189310b0dce0a5fa8b30ac94f818bcbcb"},"schema_version":"1.0"},"canonical_sha256":"3d6a4c6a47b9f9ec6418ead021ab0676ba8266903448eaa08556defbd16929a8","source":{"kind":"arxiv","id":"1802.06509","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1802.06509","created_at":"2026-05-18T00:13:44Z"},{"alias_kind":"arxiv_version","alias_value":"1802.06509v2","created_at":"2026-05-18T00:13:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.06509","created_at":"2026-05-18T00:13:44Z"},{"alias_kind":"pith_short_12","alias_value":"HVVEY2SHXH46","created_at":"2026-05-18T12:32:28Z"},{"alias_kind":"pith_short_16","alias_value":"HVVEY2SHXH46YZAY","created_at":"2026-05-18T12:32:28Z"},{"alias_kind":"pith_short_8","alias_value":"HVVEY2SH","created_at":"2026-05-18T12:32:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:HVVEY2SHXH46YZAY5LICDKYGO2","target":"record","payload":{"canonical_record":{"source":{"id":"1802.06509","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-02-19T03:08:26Z","cross_cats_sorted":[],"title_canon_sha256":"cc14b4c7de347827b8bd654b047970e2e4d3a56b9487af8e289776abd165e3fc","abstract_canon_sha256":"e9a3d464cf586fc671a7a71e44bf72d189310b0dce0a5fa8b30ac94f818bcbcb"},"schema_version":"1.0"},"canonical_sha256":"3d6a4c6a47b9f9ec6418ead021ab0676ba8266903448eaa08556defbd16929a8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:13:44.820224Z","signature_b64":"Lbol6Zu2NGhKtI1fBv3B3ZWZoVuemfccYhDBkxwHnUJK/G1AEZT6HVqR1JcIgqoa+68WTKPVzIUSitK/OYoPDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3d6a4c6a47b9f9ec6418ead021ab0676ba8266903448eaa08556defbd16929a8","last_reissued_at":"2026-05-18T00:13:44.819581Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:13:44.819581Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1802.06509","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":"2026-05-18T00:13:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gbXsHFyZPDn13c2ff/FczAuRiMnkf0Xp3yrLAoi85rlv4pBiFsyRaa56G/6liIzT/AwIA1OPnITWeNtQCMgOBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T17:31:26.061044Z"},"content_sha256":"f42354dcf336a7fa0f2762ae25550f1877de45069d9f676192af338c0734513c","schema_version":"1.0","event_id":"sha256:f42354dcf336a7fa0f2762ae25550f1877de45069d9f676192af338c0734513c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:HVVEY2SHXH46YZAY5LICDKYGO2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On the Optimization of Deep Networks: Implicit Acceleration by Overparameterization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Elad Hazan, Nadav Cohen, Sanjeev Arora","submitted_at":"2018-02-19T03:08:26Z","abstract_excerpt":"Conventional wisdom in deep learning states that increasing depth improves expressiveness but complicates optimization. This paper suggests that, sometimes, increasing depth can speed up optimization. The effect of depth on optimization is decoupled from expressiveness by focusing on settings where additional layers amount to overparameterization - linear neural networks, a well-studied model. Theoretical analysis, as well as experiments, show that here depth acts as a preconditioner which may accelerate convergence. Even on simple convex problems such as linear regression with $\\ell_p$ loss, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1802.06509","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":""},"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":"2026-05-18T00:13:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DZ1xu27zY9MnNsZCXCvUrNVpytGEjFziulZ1YH4V+JlaxAHJYL5U0lnXxYAcQh5jzmEzZIdHEnWehcYiDhmpAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T17:31:26.061891Z"},"content_sha256":"1e602ff882312255de8628f1dc41f44cbf147ccf72657bb694dbb163c6b61388","schema_version":"1.0","event_id":"sha256:1e602ff882312255de8628f1dc41f44cbf147ccf72657bb694dbb163c6b61388"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HVVEY2SHXH46YZAY5LICDKYGO2/bundle.json","state_url":"https://pith.science/pith/HVVEY2SHXH46YZAY5LICDKYGO2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HVVEY2SHXH46YZAY5LICDKYGO2/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":"2026-08-03T17:31:26Z","links":{"resolver":"https://pith.science/pith/HVVEY2SHXH46YZAY5LICDKYGO2","bundle":"https://pith.science/pith/HVVEY2SHXH46YZAY5LICDKYGO2/bundle.json","state":"https://pith.science/pith/HVVEY2SHXH46YZAY5LICDKYGO2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HVVEY2SHXH46YZAY5LICDKYGO2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:HVVEY2SHXH46YZAY5LICDKYGO2","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":"e9a3d464cf586fc671a7a71e44bf72d189310b0dce0a5fa8b30ac94f818bcbcb","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-02-19T03:08:26Z","title_canon_sha256":"cc14b4c7de347827b8bd654b047970e2e4d3a56b9487af8e289776abd165e3fc"},"schema_version":"1.0","source":{"id":"1802.06509","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1802.06509","created_at":"2026-05-18T00:13:44Z"},{"alias_kind":"arxiv_version","alias_value":"1802.06509v2","created_at":"2026-05-18T00:13:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.06509","created_at":"2026-05-18T00:13:44Z"},{"alias_kind":"pith_short_12","alias_value":"HVVEY2SHXH46","created_at":"2026-05-18T12:32:28Z"},{"alias_kind":"pith_short_16","alias_value":"HVVEY2SHXH46YZAY","created_at":"2026-05-18T12:32:28Z"},{"alias_kind":"pith_short_8","alias_value":"HVVEY2SH","created_at":"2026-05-18T12:32:28Z"}],"graph_snapshots":[{"event_id":"sha256:1e602ff882312255de8628f1dc41f44cbf147ccf72657bb694dbb163c6b61388","target":"graph","created_at":"2026-05-18T00:13:44Z","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"},"paper":{"abstract_excerpt":"Conventional wisdom in deep learning states that increasing depth improves expressiveness but complicates optimization. This paper suggests that, sometimes, increasing depth can speed up optimization. The effect of depth on optimization is decoupled from expressiveness by focusing on settings where additional layers amount to overparameterization - linear neural networks, a well-studied model. Theoretical analysis, as well as experiments, show that here depth acts as a preconditioner which may accelerate convergence. Even on simple convex problems such as linear regression with $\\ell_p$ loss, ","authors_text":"Elad Hazan, Nadav Cohen, Sanjeev Arora","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-02-19T03:08:26Z","title":"On the Optimization of Deep Networks: Implicit Acceleration by Overparameterization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1802.06509","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:f42354dcf336a7fa0f2762ae25550f1877de45069d9f676192af338c0734513c","target":"record","created_at":"2026-05-18T00:13:44Z","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":"e9a3d464cf586fc671a7a71e44bf72d189310b0dce0a5fa8b30ac94f818bcbcb","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-02-19T03:08:26Z","title_canon_sha256":"cc14b4c7de347827b8bd654b047970e2e4d3a56b9487af8e289776abd165e3fc"},"schema_version":"1.0","source":{"id":"1802.06509","kind":"arxiv","version":2}},"canonical_sha256":"3d6a4c6a47b9f9ec6418ead021ab0676ba8266903448eaa08556defbd16929a8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3d6a4c6a47b9f9ec6418ead021ab0676ba8266903448eaa08556defbd16929a8","first_computed_at":"2026-05-18T00:13:44.819581Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:13:44.819581Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Lbol6Zu2NGhKtI1fBv3B3ZWZoVuemfccYhDBkxwHnUJK/G1AEZT6HVqR1JcIgqoa+68WTKPVzIUSitK/OYoPDQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:13:44.820224Z","signed_message":"canonical_sha256_bytes"},"source_id":"1802.06509","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f42354dcf336a7fa0f2762ae25550f1877de45069d9f676192af338c0734513c","sha256:1e602ff882312255de8628f1dc41f44cbf147ccf72657bb694dbb163c6b61388"],"state_sha256":"a2d8ad5519ef57221673479828d45434024923517d89b349af6ccc320832d422"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g5ElToTzyLLgfsMzzZEWf5TgRNb7BAfdUhCMRArWOi+5Z4KlWHa996FFlKJWYLc05hYrNJHKiE5Wy+kZOum2DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T17:31:26.069731Z","bundle_sha256":"5d32e4ef04a53b7f10e420f205adda75f0e2ff36b2d7fd3fc692b29c8453e644"}}