{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:77FB2DPW2WD442GRCSFYNYT3UU","short_pith_number":"pith:77FB2DPW","schema_version":"1.0","canonical_sha256":"ffca1d0df6d587ce68d1148b86e27ba53fe50b2f735eb0b6725e83c8116ef8cc","source":{"kind":"arxiv","id":"1810.01973","version":1},"attestation_state":"computed","paper":{"title":"Sparse Winograd Convolutional neural networks on small-scale systolic arrays","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.DC","authors_text":"Benjamin Kuschner, Feng Shi, Haochen Li, Song-Chun Zhu, Yuhe Gao","submitted_at":"2018-10-03T21:01:51Z","abstract_excerpt":"The reconfigurability, energy-efficiency, and massive parallelism on FPGAs make them one of the best choices for implementing efficient deep learning accelerators. However, state-of-art implementations seldom consider the balance between high throughput of computation power and the ability of the memory subsystem to support it. In this paper, we implement an accelerator on FPGA by combining the sparse Winograd convolution, clusters of small-scale systolic arrays, and a tailored memory layout design. We also provide an analytical model analysis for the general Winograd convolution algorithm as "},"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":"1810.01973","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2018-10-03T21:01:51Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"6f81689990b896dc02bca057f4e2e163d065163057de40f54fed6625842981e0","abstract_canon_sha256":"56306e5f983da402853c3c93d13dded040e12113e86838cd40e690f06ed7088f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:04:06.714389Z","signature_b64":"z/3ACXI8qt2HvlgmetWFx285qJG3LxfJh/eQd8iwxi+xYBgtqvIqYLfnRfFgV7c6GKnNp/uYgVL2NJ2NQeMBDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ffca1d0df6d587ce68d1148b86e27ba53fe50b2f735eb0b6725e83c8116ef8cc","last_reissued_at":"2026-05-18T00:04:06.713695Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:04:06.713695Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sparse Winograd Convolutional neural networks on small-scale systolic arrays","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.DC","authors_text":"Benjamin Kuschner, Feng Shi, Haochen Li, Song-Chun Zhu, Yuhe Gao","submitted_at":"2018-10-03T21:01:51Z","abstract_excerpt":"The reconfigurability, energy-efficiency, and massive parallelism on FPGAs make them one of the best choices for implementing efficient deep learning accelerators. However, state-of-art implementations seldom consider the balance between high throughput of computation power and the ability of the memory subsystem to support it. In this paper, we implement an accelerator on FPGA by combining the sparse Winograd convolution, clusters of small-scale systolic arrays, and a tailored memory layout design. We also provide an analytical model analysis for the general Winograd convolution algorithm as "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.01973","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":"1810.01973","created_at":"2026-05-18T00:04:06.713780+00:00"},{"alias_kind":"arxiv_version","alias_value":"1810.01973v1","created_at":"2026-05-18T00:04:06.713780+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.01973","created_at":"2026-05-18T00:04:06.713780+00:00"},{"alias_kind":"pith_short_12","alias_value":"77FB2DPW2WD4","created_at":"2026-05-18T12:32:11.075285+00:00"},{"alias_kind":"pith_short_16","alias_value":"77FB2DPW2WD442GR","created_at":"2026-05-18T12:32:11.075285+00:00"},{"alias_kind":"pith_short_8","alias_value":"77FB2DPW","created_at":"2026-05-18T12:32:11.075285+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/77FB2DPW2WD442GRCSFYNYT3UU","json":"https://pith.science/pith/77FB2DPW2WD442GRCSFYNYT3UU.json","graph_json":"https://pith.science/api/pith-number/77FB2DPW2WD442GRCSFYNYT3UU/graph.json","events_json":"https://pith.science/api/pith-number/77FB2DPW2WD442GRCSFYNYT3UU/events.json","paper":"https://pith.science/paper/77FB2DPW"},"agent_actions":{"view_html":"https://pith.science/pith/77FB2DPW2WD442GRCSFYNYT3UU","download_json":"https://pith.science/pith/77FB2DPW2WD442GRCSFYNYT3UU.json","view_paper":"https://pith.science/paper/77FB2DPW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1810.01973&json=true","fetch_graph":"https://pith.science/api/pith-number/77FB2DPW2WD442GRCSFYNYT3UU/graph.json","fetch_events":"https://pith.science/api/pith-number/77FB2DPW2WD442GRCSFYNYT3UU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/77FB2DPW2WD442GRCSFYNYT3UU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/77FB2DPW2WD442GRCSFYNYT3UU/action/storage_attestation","attest_author":"https://pith.science/pith/77FB2DPW2WD442GRCSFYNYT3UU/action/author_attestation","sign_citation":"https://pith.science/pith/77FB2DPW2WD442GRCSFYNYT3UU/action/citation_signature","submit_replication":"https://pith.science/pith/77FB2DPW2WD442GRCSFYNYT3UU/action/replication_record"}},"created_at":"2026-05-18T00:04:06.713780+00:00","updated_at":"2026-05-18T00:04:06.713780+00:00"}