{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:ZF52CYP2IHOMAHVPKY2OIK3TQO","short_pith_number":"pith:ZF52CYP2","schema_version":"1.0","canonical_sha256":"c97ba161fa41dcc01eaf5634e42b7383a7adc5d2140fb0663535232a8549cc55","source":{"kind":"arxiv","id":"1809.07851","version":1},"attestation_state":"computed","paper":{"title":"FFT Convolutions are Faster than Winograd on Modern CPUs, Here is Why","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.PF","authors_text":"Aleksandar Zlateski, Fredo Durand, Kai Li, Zhen Jia","submitted_at":"2018-09-20T20:55:15Z","abstract_excerpt":"Winograd-based convolution has quickly gained traction as a preferred approach to implement convolutional neural networks (ConvNet) on various hardware platforms because it requires fewer floating point operations than FFT-based or direct convolutions.\n  This paper compares three highly optimized implementations (regular FFT--, Gauss--FFT--, and Winograd--based convolutions) on modern multi-- and many--core CPUs. Although all three implementations employed the same optimizations for modern CPUs, our experimental results with two popular ConvNets (VGG and AlexNet) show that the FFT--based imple"},"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":"1809.07851","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.PF","submitted_at":"2018-09-20T20:55:15Z","cross_cats_sorted":[],"title_canon_sha256":"eeebc96d089d07658f4d5a0e243269a401590a7c37304a5d26726143c25fc14a","abstract_canon_sha256":"c9bd16132fdc490796661679674a16c05fc92ebd31bbafa21c194059937ee1be"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:05:11.125762Z","signature_b64":"HJqHOMU2tuYSBOHbSfwI56Hhfhl4fX8rwbv6jkeEtaJh4OUnQ6M93SCxilmaMu2GVAohwhNyIbLj8xUgdj7JDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c97ba161fa41dcc01eaf5634e42b7383a7adc5d2140fb0663535232a8549cc55","last_reissued_at":"2026-05-18T00:05:11.125296Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:05:11.125296Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FFT Convolutions are Faster than Winograd on Modern CPUs, Here is Why","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.PF","authors_text":"Aleksandar Zlateski, Fredo Durand, Kai Li, Zhen Jia","submitted_at":"2018-09-20T20:55:15Z","abstract_excerpt":"Winograd-based convolution has quickly gained traction as a preferred approach to implement convolutional neural networks (ConvNet) on various hardware platforms because it requires fewer floating point operations than FFT-based or direct convolutions.\n  This paper compares three highly optimized implementations (regular FFT--, Gauss--FFT--, and Winograd--based convolutions) on modern multi-- and many--core CPUs. Although all three implementations employed the same optimizations for modern CPUs, our experimental results with two popular ConvNets (VGG and AlexNet) show that the FFT--based imple"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1809.07851","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":""},"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":"1809.07851","created_at":"2026-05-18T00:05:11.125388+00:00"},{"alias_kind":"arxiv_version","alias_value":"1809.07851v1","created_at":"2026-05-18T00:05:11.125388+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1809.07851","created_at":"2026-05-18T00:05:11.125388+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZF52CYP2IHOM","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZF52CYP2IHOMAHVP","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZF52CYP2","created_at":"2026-05-18T12:33:07.085635+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/ZF52CYP2IHOMAHVPKY2OIK3TQO","json":"https://pith.science/pith/ZF52CYP2IHOMAHVPKY2OIK3TQO.json","graph_json":"https://pith.science/api/pith-number/ZF52CYP2IHOMAHVPKY2OIK3TQO/graph.json","events_json":"https://pith.science/api/pith-number/ZF52CYP2IHOMAHVPKY2OIK3TQO/events.json","paper":"https://pith.science/paper/ZF52CYP2"},"agent_actions":{"view_html":"https://pith.science/pith/ZF52CYP2IHOMAHVPKY2OIK3TQO","download_json":"https://pith.science/pith/ZF52CYP2IHOMAHVPKY2OIK3TQO.json","view_paper":"https://pith.science/paper/ZF52CYP2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1809.07851&json=true","fetch_graph":"https://pith.science/api/pith-number/ZF52CYP2IHOMAHVPKY2OIK3TQO/graph.json","fetch_events":"https://pith.science/api/pith-number/ZF52CYP2IHOMAHVPKY2OIK3TQO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZF52CYP2IHOMAHVPKY2OIK3TQO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZF52CYP2IHOMAHVPKY2OIK3TQO/action/storage_attestation","attest_author":"https://pith.science/pith/ZF52CYP2IHOMAHVPKY2OIK3TQO/action/author_attestation","sign_citation":"https://pith.science/pith/ZF52CYP2IHOMAHVPKY2OIK3TQO/action/citation_signature","submit_replication":"https://pith.science/pith/ZF52CYP2IHOMAHVPKY2OIK3TQO/action/replication_record"}},"created_at":"2026-05-18T00:05:11.125388+00:00","updated_at":"2026-05-18T00:05:11.125388+00:00"}