{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:AIDIAUPQJSYVHOLDRJ3PD3GOUP","short_pith_number":"pith:AIDIAUPQ","schema_version":"1.0","canonical_sha256":"02068051f04cb153b9638a76f1eccea3d3ec73e5bf8796b56c14af12657bf691","source":{"kind":"arxiv","id":"1709.01134","version":1},"attestation_state":"computed","paper":{"title":"WRPN: Wide Reduced-Precision Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.NE"],"primary_cat":"cs.CV","authors_text":"Asit Mishra, Debbie Marr, Eriko Nurvitadhi, Jeffrey J Cook","submitted_at":"2017-09-04T19:56:48Z","abstract_excerpt":"For computer vision applications, prior works have shown the efficacy of reducing numeric precision of model parameters (network weights) in deep neural networks. Activation maps, however, occupy a large memory footprint during both the training and inference step when using mini-batches of inputs. One way to reduce this large memory footprint is to reduce the precision of activations. However, past works have shown that reducing the precision of activations hurts model accuracy. We study schemes to train networks from scratch using reduced-precision activations without hurting accuracy. We re"},"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":"1709.01134","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2017-09-04T19:56:48Z","cross_cats_sorted":["cs.LG","cs.NE"],"title_canon_sha256":"7b8625fac6839e1fe5b29caa660685ac8e1cbdc981146130d1141cb51ff3c5b0","abstract_canon_sha256":"b94b8f5438b2c766a0e16d702985b36bf63721da98315a5e1a5718bd5c85133e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:36:00.827721Z","signature_b64":"V5YKYhYcVnxHW1rkrPPNjqet9nEhs4ujKWbwxkusvDmzAT+bdCu/206sBYRQaIeDiQNNBdkZkpOnDbycx3QGDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"02068051f04cb153b9638a76f1eccea3d3ec73e5bf8796b56c14af12657bf691","last_reissued_at":"2026-05-18T00:36:00.827090Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:36:00.827090Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"WRPN: Wide Reduced-Precision Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.NE"],"primary_cat":"cs.CV","authors_text":"Asit Mishra, Debbie Marr, Eriko Nurvitadhi, Jeffrey J Cook","submitted_at":"2017-09-04T19:56:48Z","abstract_excerpt":"For computer vision applications, prior works have shown the efficacy of reducing numeric precision of model parameters (network weights) in deep neural networks. Activation maps, however, occupy a large memory footprint during both the training and inference step when using mini-batches of inputs. One way to reduce this large memory footprint is to reduce the precision of activations. However, past works have shown that reducing the precision of activations hurts model accuracy. We study schemes to train networks from scratch using reduced-precision activations without hurting accuracy. We re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1709.01134","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":"1709.01134","created_at":"2026-05-18T00:36:00.827183+00:00"},{"alias_kind":"arxiv_version","alias_value":"1709.01134v1","created_at":"2026-05-18T00:36:00.827183+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1709.01134","created_at":"2026-05-18T00:36:00.827183+00:00"},{"alias_kind":"pith_short_12","alias_value":"AIDIAUPQJSYV","created_at":"2026-05-18T12:31:05.417338+00:00"},{"alias_kind":"pith_short_16","alias_value":"AIDIAUPQJSYVHOLD","created_at":"2026-05-18T12:31:05.417338+00:00"},{"alias_kind":"pith_short_8","alias_value":"AIDIAUPQ","created_at":"2026-05-18T12:31:05.417338+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"1710.03740","citing_title":"Mixed Precision Training","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AIDIAUPQJSYVHOLDRJ3PD3GOUP","json":"https://pith.science/pith/AIDIAUPQJSYVHOLDRJ3PD3GOUP.json","graph_json":"https://pith.science/api/pith-number/AIDIAUPQJSYVHOLDRJ3PD3GOUP/graph.json","events_json":"https://pith.science/api/pith-number/AIDIAUPQJSYVHOLDRJ3PD3GOUP/events.json","paper":"https://pith.science/paper/AIDIAUPQ"},"agent_actions":{"view_html":"https://pith.science/pith/AIDIAUPQJSYVHOLDRJ3PD3GOUP","download_json":"https://pith.science/pith/AIDIAUPQJSYVHOLDRJ3PD3GOUP.json","view_paper":"https://pith.science/paper/AIDIAUPQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1709.01134&json=true","fetch_graph":"https://pith.science/api/pith-number/AIDIAUPQJSYVHOLDRJ3PD3GOUP/graph.json","fetch_events":"https://pith.science/api/pith-number/AIDIAUPQJSYVHOLDRJ3PD3GOUP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AIDIAUPQJSYVHOLDRJ3PD3GOUP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AIDIAUPQJSYVHOLDRJ3PD3GOUP/action/storage_attestation","attest_author":"https://pith.science/pith/AIDIAUPQJSYVHOLDRJ3PD3GOUP/action/author_attestation","sign_citation":"https://pith.science/pith/AIDIAUPQJSYVHOLDRJ3PD3GOUP/action/citation_signature","submit_replication":"https://pith.science/pith/AIDIAUPQJSYVHOLDRJ3PD3GOUP/action/replication_record"}},"created_at":"2026-05-18T00:36:00.827183+00:00","updated_at":"2026-05-18T00:36:00.827183+00:00"}