{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:F7BFWG2N45WRLP2AZGV3K2R64V","short_pith_number":"pith:F7BFWG2N","schema_version":"1.0","canonical_sha256":"2fc25b1b4de76d15bf40c9abb56a3ee552562eeff4f9c1df3ecde2abfcbd31d9","source":{"kind":"arxiv","id":"2110.04622","version":1},"attestation_state":"computed","paper":{"title":"Does Preprocessing Help Training Over-parameterized Neural Networks?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DS","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ruizhe Zhang, Shuo Yang, Zhao Song","submitted_at":"2021-10-09T18:16:23Z","abstract_excerpt":"Deep neural networks have achieved impressive performance in many areas. Designing a fast and provable method for training neural networks is a fundamental question in machine learning.\n  The classical training method requires paying $\\Omega(mnd)$ cost for both forward computation and backward computation, where $m$ is the width of the neural network, and we are given $n$ training points in $d$-dimensional space. In this paper, we propose two novel preprocessing ideas to bypass this $\\Omega(mnd)$ barrier:\n  $\\bullet$ First, by preprocessing the initial weights of the neural networks, we can tr"},"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":"2110.04622","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-09T18:16:23Z","cross_cats_sorted":["cs.DS","stat.ML"],"title_canon_sha256":"f881ea4fee2beba01233be7ac841ac16b281a63019efa31e949caacc72903c38","abstract_canon_sha256":"f22a8a3b72262d366b9fba8f1380a58b74bbb00f70a4696c2a20fd5766350884"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:21:19.391633Z","signature_b64":"o8nrIImHAqNmk9l6bIaogXu7EA/aCgeDJqDwUHDuAQ2sT/AaHG75o0sisf/yAWUt0LgJf3Om9FFjiM4z54PBCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2fc25b1b4de76d15bf40c9abb56a3ee552562eeff4f9c1df3ecde2abfcbd31d9","last_reissued_at":"2026-07-05T03:21:19.391209Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:21:19.391209Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Does Preprocessing Help Training Over-parameterized Neural Networks?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DS","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ruizhe Zhang, Shuo Yang, Zhao Song","submitted_at":"2021-10-09T18:16:23Z","abstract_excerpt":"Deep neural networks have achieved impressive performance in many areas. Designing a fast and provable method for training neural networks is a fundamental question in machine learning.\n  The classical training method requires paying $\\Omega(mnd)$ cost for both forward computation and backward computation, where $m$ is the width of the neural network, and we are given $n$ training points in $d$-dimensional space. In this paper, we propose two novel preprocessing ideas to bypass this $\\Omega(mnd)$ barrier:\n  $\\bullet$ First, by preprocessing the initial weights of the neural networks, we can tr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.04622","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/2110.04622/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":"2110.04622","created_at":"2026-07-05T03:21:19.391271+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.04622v1","created_at":"2026-07-05T03:21:19.391271+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.04622","created_at":"2026-07-05T03:21:19.391271+00:00"},{"alias_kind":"pith_short_12","alias_value":"F7BFWG2N45WR","created_at":"2026-07-05T03:21:19.391271+00:00"},{"alias_kind":"pith_short_16","alias_value":"F7BFWG2N45WRLP2A","created_at":"2026-07-05T03:21:19.391271+00:00"},{"alias_kind":"pith_short_8","alias_value":"F7BFWG2N","created_at":"2026-07-05T03:21:19.391271+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F7BFWG2N45WRLP2AZGV3K2R64V","json":"https://pith.science/pith/F7BFWG2N45WRLP2AZGV3K2R64V.json","graph_json":"https://pith.science/api/pith-number/F7BFWG2N45WRLP2AZGV3K2R64V/graph.json","events_json":"https://pith.science/api/pith-number/F7BFWG2N45WRLP2AZGV3K2R64V/events.json","paper":"https://pith.science/paper/F7BFWG2N"},"agent_actions":{"view_html":"https://pith.science/pith/F7BFWG2N45WRLP2AZGV3K2R64V","download_json":"https://pith.science/pith/F7BFWG2N45WRLP2AZGV3K2R64V.json","view_paper":"https://pith.science/paper/F7BFWG2N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.04622&json=true","fetch_graph":"https://pith.science/api/pith-number/F7BFWG2N45WRLP2AZGV3K2R64V/graph.json","fetch_events":"https://pith.science/api/pith-number/F7BFWG2N45WRLP2AZGV3K2R64V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F7BFWG2N45WRLP2AZGV3K2R64V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F7BFWG2N45WRLP2AZGV3K2R64V/action/storage_attestation","attest_author":"https://pith.science/pith/F7BFWG2N45WRLP2AZGV3K2R64V/action/author_attestation","sign_citation":"https://pith.science/pith/F7BFWG2N45WRLP2AZGV3K2R64V/action/citation_signature","submit_replication":"https://pith.science/pith/F7BFWG2N45WRLP2AZGV3K2R64V/action/replication_record"}},"created_at":"2026-07-05T03:21:19.391271+00:00","updated_at":"2026-07-05T03:21:19.391271+00:00"}