{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:SWUUAVRTXMPLHZRTH4ITN2OBNR","short_pith_number":"pith:SWUUAVRT","schema_version":"1.0","canonical_sha256":"95a9405633bb1eb3e6333f1136e9c16c77f70bad6a657137737b92582b4470bf","source":{"kind":"arxiv","id":"1901.06588","version":1},"attestation_state":"computed","paper":{"title":"Accumulation Bit-Width Scaling For Ultra-Low Precision Training Of Deep Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ankur Agrawal, Charbel Sakr, Chia-Yu Chen, Jungwook Choi, Kailash Gopalakrishnan, Naigang Wang, Naresh Shanbhag","submitted_at":"2019-01-19T21:27:28Z","abstract_excerpt":"Efforts to reduce the numerical precision of computations in deep learning training have yielded systems that aggressively quantize weights and activations, yet employ wide high-precision accumulators for partial sums in inner-product operations to preserve the quality of convergence. The absence of any framework to analyze the precision requirements of partial sum accumulations results in conservative design choices. This imposes an upper-bound on the reduction of complexity of multiply-accumulate units. We present a statistical approach to analyze the impact of reduced accumulation precision"},"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":"1901.06588","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-19T21:27:28Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"7439829f09ba2a90cb46a452a19396d46e116985350c118e03370ba1aa6679ea","abstract_canon_sha256":"7390be3b622977a1dd6bd71535bd273bfd28111c4486d26ac1cbb2bb33063c3a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:55:49.726270Z","signature_b64":"Bdxs6NwjQGlo1a9iCoC3BCcP/nWm8wA1N8C9LIt1zQuDTnHlnTH9uukZsKyY+lcJfBKO9Hvh6TZoAuuX+h+/CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"95a9405633bb1eb3e6333f1136e9c16c77f70bad6a657137737b92582b4470bf","last_reissued_at":"2026-05-17T23:55:49.725594Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:55:49.725594Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Accumulation Bit-Width Scaling For Ultra-Low Precision Training Of Deep Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ankur Agrawal, Charbel Sakr, Chia-Yu Chen, Jungwook Choi, Kailash Gopalakrishnan, Naigang Wang, Naresh Shanbhag","submitted_at":"2019-01-19T21:27:28Z","abstract_excerpt":"Efforts to reduce the numerical precision of computations in deep learning training have yielded systems that aggressively quantize weights and activations, yet employ wide high-precision accumulators for partial sums in inner-product operations to preserve the quality of convergence. The absence of any framework to analyze the precision requirements of partial sum accumulations results in conservative design choices. This imposes an upper-bound on the reduction of complexity of multiply-accumulate units. We present a statistical approach to analyze the impact of reduced accumulation precision"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.06588","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":"1901.06588","created_at":"2026-05-17T23:55:49.725701+00:00"},{"alias_kind":"arxiv_version","alias_value":"1901.06588v1","created_at":"2026-05-17T23:55:49.725701+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.06588","created_at":"2026-05-17T23:55:49.725701+00:00"},{"alias_kind":"pith_short_12","alias_value":"SWUUAVRTXMPL","created_at":"2026-05-18T12:33:27.125529+00:00"},{"alias_kind":"pith_short_16","alias_value":"SWUUAVRTXMPLHZRT","created_at":"2026-05-18T12:33:27.125529+00:00"},{"alias_kind":"pith_short_8","alias_value":"SWUUAVRT","created_at":"2026-05-18T12:33:27.125529+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/SWUUAVRTXMPLHZRTH4ITN2OBNR","json":"https://pith.science/pith/SWUUAVRTXMPLHZRTH4ITN2OBNR.json","graph_json":"https://pith.science/api/pith-number/SWUUAVRTXMPLHZRTH4ITN2OBNR/graph.json","events_json":"https://pith.science/api/pith-number/SWUUAVRTXMPLHZRTH4ITN2OBNR/events.json","paper":"https://pith.science/paper/SWUUAVRT"},"agent_actions":{"view_html":"https://pith.science/pith/SWUUAVRTXMPLHZRTH4ITN2OBNR","download_json":"https://pith.science/pith/SWUUAVRTXMPLHZRTH4ITN2OBNR.json","view_paper":"https://pith.science/paper/SWUUAVRT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1901.06588&json=true","fetch_graph":"https://pith.science/api/pith-number/SWUUAVRTXMPLHZRTH4ITN2OBNR/graph.json","fetch_events":"https://pith.science/api/pith-number/SWUUAVRTXMPLHZRTH4ITN2OBNR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SWUUAVRTXMPLHZRTH4ITN2OBNR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SWUUAVRTXMPLHZRTH4ITN2OBNR/action/storage_attestation","attest_author":"https://pith.science/pith/SWUUAVRTXMPLHZRTH4ITN2OBNR/action/author_attestation","sign_citation":"https://pith.science/pith/SWUUAVRTXMPLHZRTH4ITN2OBNR/action/citation_signature","submit_replication":"https://pith.science/pith/SWUUAVRTXMPLHZRTH4ITN2OBNR/action/replication_record"}},"created_at":"2026-05-17T23:55:49.725701+00:00","updated_at":"2026-05-17T23:55:49.725701+00:00"}