{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:I5RSTK3P4BIXPDHSF5FGEY4TEP","short_pith_number":"pith:I5RSTK3P","schema_version":"1.0","canonical_sha256":"476329ab6fe051778cf22f4a62639323c7592e524420f9ac230bfe2ae307fe2a","source":{"kind":"arxiv","id":"1812.04353","version":3},"attestation_state":"computed","paper":{"title":"Proximal Mean-field for Neural Network Quantization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Philip H. S. Torr, Puneet K. Dokania, Richard Hartley, Thalaiyasingam Ajanthan","submitted_at":"2018-12-11T12:27:54Z","abstract_excerpt":"Compressing large Neural Networks (NN) by quantizing the parameters, while maintaining the performance is highly desirable due to reduced memory and time complexity. In this work, we cast NN quantization as a discrete labelling problem, and by examining relaxations, we design an efficient iterative optimization procedure that involves stochastic gradient descent followed by a projection. We prove that our simple projected gradient descent approach is, in fact, equivalent to a proximal version of the well-known mean-field method. These findings would allow the decades-old and theoretically grou"},"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":"1812.04353","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-12-11T12:27:54Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"29722f1157acb0573c94c23663d5ded1499c15ae2856e8f26794719267d6a8f9","abstract_canon_sha256":"fca3fce26e8f2edce7781fb9d55c65f7b312c25cf43faac882f10b51baa826a3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:58:24.418284Z","signature_b64":"YagY9U6hPLxZ19djqPsA2GegEaWqSh0cW2IJ8H+9gewQ+RlQ7v1kVBKe5T15sSTHGw0W7Bvll5WVk+iuePxtBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"476329ab6fe051778cf22f4a62639323c7592e524420f9ac230bfe2ae307fe2a","last_reissued_at":"2026-07-04T23:58:24.417829Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:58:24.417829Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Proximal Mean-field for Neural Network Quantization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Philip H. S. Torr, Puneet K. Dokania, Richard Hartley, Thalaiyasingam Ajanthan","submitted_at":"2018-12-11T12:27:54Z","abstract_excerpt":"Compressing large Neural Networks (NN) by quantizing the parameters, while maintaining the performance is highly desirable due to reduced memory and time complexity. In this work, we cast NN quantization as a discrete labelling problem, and by examining relaxations, we design an efficient iterative optimization procedure that involves stochastic gradient descent followed by a projection. We prove that our simple projected gradient descent approach is, in fact, equivalent to a proximal version of the well-known mean-field method. These findings would allow the decades-old and theoretically grou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1812.04353","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1812.04353/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":"1812.04353","created_at":"2026-07-04T23:58:24.417885+00:00"},{"alias_kind":"arxiv_version","alias_value":"1812.04353v3","created_at":"2026-07-04T23:58:24.417885+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1812.04353","created_at":"2026-07-04T23:58:24.417885+00:00"},{"alias_kind":"pith_short_12","alias_value":"I5RSTK3P4BIX","created_at":"2026-07-04T23:58:24.417885+00:00"},{"alias_kind":"pith_short_16","alias_value":"I5RSTK3P4BIXPDHS","created_at":"2026-07-04T23:58:24.417885+00:00"},{"alias_kind":"pith_short_8","alias_value":"I5RSTK3P","created_at":"2026-07-04T23:58:24.417885+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/I5RSTK3P4BIXPDHSF5FGEY4TEP","json":"https://pith.science/pith/I5RSTK3P4BIXPDHSF5FGEY4TEP.json","graph_json":"https://pith.science/api/pith-number/I5RSTK3P4BIXPDHSF5FGEY4TEP/graph.json","events_json":"https://pith.science/api/pith-number/I5RSTK3P4BIXPDHSF5FGEY4TEP/events.json","paper":"https://pith.science/paper/I5RSTK3P"},"agent_actions":{"view_html":"https://pith.science/pith/I5RSTK3P4BIXPDHSF5FGEY4TEP","download_json":"https://pith.science/pith/I5RSTK3P4BIXPDHSF5FGEY4TEP.json","view_paper":"https://pith.science/paper/I5RSTK3P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1812.04353&json=true","fetch_graph":"https://pith.science/api/pith-number/I5RSTK3P4BIXPDHSF5FGEY4TEP/graph.json","fetch_events":"https://pith.science/api/pith-number/I5RSTK3P4BIXPDHSF5FGEY4TEP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I5RSTK3P4BIXPDHSF5FGEY4TEP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I5RSTK3P4BIXPDHSF5FGEY4TEP/action/storage_attestation","attest_author":"https://pith.science/pith/I5RSTK3P4BIXPDHSF5FGEY4TEP/action/author_attestation","sign_citation":"https://pith.science/pith/I5RSTK3P4BIXPDHSF5FGEY4TEP/action/citation_signature","submit_replication":"https://pith.science/pith/I5RSTK3P4BIXPDHSF5FGEY4TEP/action/replication_record"}},"created_at":"2026-07-04T23:58:24.417885+00:00","updated_at":"2026-07-04T23:58:24.417885+00:00"}