{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:OL6ATQWIILOQQNM4HV74F5TRBT","short_pith_number":"pith:OL6ATQWI","schema_version":"1.0","canonical_sha256":"72fc09c2c842dd08359c3d7fc2f6710cf2529d8b94c620b10b70a44f7b80aa90","source":{"kind":"arxiv","id":"1701.08978","version":2},"attestation_state":"computed","paper":{"title":"Mixed Low-precision Deep Learning Inference using Dynamic Fixed Point","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE"],"primary_cat":"cs.LG","authors_text":"Abhisek Kundu, Bharat Kaul, Dheevatsa Mudigere, Dipankar Das, Naveen Mellempudi","submitted_at":"2017-01-31T10:28:37Z","abstract_excerpt":"We propose a cluster-based quantization method to convert pre-trained full precision weights into ternary weights with minimal impact on the accuracy. In addition, we also constrain the activations to 8-bits thus enabling sub 8-bit full integer inference pipeline. Our method uses smaller clusters of N filters with a common scaling factor to minimize the quantization loss, while also maximizing the number of ternary operations. We show that with a cluster size of N=4 on Resnet-101, can achieve 71.8% TOP-1 accuracy, within 6% of the best full precision results while replacing ~85% of all multipl"},"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":"1701.08978","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-01-31T10:28:37Z","cross_cats_sorted":["cs.NE"],"title_canon_sha256":"d7c6abe5b30766cc16891780f3c09227799fbc20d7687136c9745b0f60af5ff7","abstract_canon_sha256":"6e3b9ca68d4e7d013995421e7b3ffe7a0d6da8457bbbe1fa4b7b8ddf2425dd97"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:51:35.415311Z","signature_b64":"edJI2F28xXvWJnCzBKgiRqfBrJIhqI93PhL6NQ+plnFdN0qgBwEQWYG5xn87wWw+1H4+e4IdM1pcKgLhabVwAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"72fc09c2c842dd08359c3d7fc2f6710cf2529d8b94c620b10b70a44f7b80aa90","last_reissued_at":"2026-05-18T00:51:35.414899Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:51:35.414899Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mixed Low-precision Deep Learning Inference using Dynamic Fixed Point","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE"],"primary_cat":"cs.LG","authors_text":"Abhisek Kundu, Bharat Kaul, Dheevatsa Mudigere, Dipankar Das, Naveen Mellempudi","submitted_at":"2017-01-31T10:28:37Z","abstract_excerpt":"We propose a cluster-based quantization method to convert pre-trained full precision weights into ternary weights with minimal impact on the accuracy. In addition, we also constrain the activations to 8-bits thus enabling sub 8-bit full integer inference pipeline. Our method uses smaller clusters of N filters with a common scaling factor to minimize the quantization loss, while also maximizing the number of ternary operations. We show that with a cluster size of N=4 on Resnet-101, can achieve 71.8% TOP-1 accuracy, within 6% of the best full precision results while replacing ~85% of all multipl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1701.08978","kind":"arxiv","version":2},"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":"1701.08978","created_at":"2026-05-18T00:51:35.414969+00:00"},{"alias_kind":"arxiv_version","alias_value":"1701.08978v2","created_at":"2026-05-18T00:51:35.414969+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1701.08978","created_at":"2026-05-18T00:51:35.414969+00:00"},{"alias_kind":"pith_short_12","alias_value":"OL6ATQWIILOQ","created_at":"2026-05-18T12:31:34.259226+00:00"},{"alias_kind":"pith_short_16","alias_value":"OL6ATQWIILOQQNM4","created_at":"2026-05-18T12:31:34.259226+00:00"},{"alias_kind":"pith_short_8","alias_value":"OL6ATQWI","created_at":"2026-05-18T12:31:34.259226+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/OL6ATQWIILOQQNM4HV74F5TRBT","json":"https://pith.science/pith/OL6ATQWIILOQQNM4HV74F5TRBT.json","graph_json":"https://pith.science/api/pith-number/OL6ATQWIILOQQNM4HV74F5TRBT/graph.json","events_json":"https://pith.science/api/pith-number/OL6ATQWIILOQQNM4HV74F5TRBT/events.json","paper":"https://pith.science/paper/OL6ATQWI"},"agent_actions":{"view_html":"https://pith.science/pith/OL6ATQWIILOQQNM4HV74F5TRBT","download_json":"https://pith.science/pith/OL6ATQWIILOQQNM4HV74F5TRBT.json","view_paper":"https://pith.science/paper/OL6ATQWI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1701.08978&json=true","fetch_graph":"https://pith.science/api/pith-number/OL6ATQWIILOQQNM4HV74F5TRBT/graph.json","fetch_events":"https://pith.science/api/pith-number/OL6ATQWIILOQQNM4HV74F5TRBT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OL6ATQWIILOQQNM4HV74F5TRBT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OL6ATQWIILOQQNM4HV74F5TRBT/action/storage_attestation","attest_author":"https://pith.science/pith/OL6ATQWIILOQQNM4HV74F5TRBT/action/author_attestation","sign_citation":"https://pith.science/pith/OL6ATQWIILOQQNM4HV74F5TRBT/action/citation_signature","submit_replication":"https://pith.science/pith/OL6ATQWIILOQQNM4HV74F5TRBT/action/replication_record"}},"created_at":"2026-05-18T00:51:35.414969+00:00","updated_at":"2026-05-18T00:51:35.414969+00:00"}