{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2013:QNC5U7N7YHMNYHWDUDWSMMZXNT","short_pith_number":"pith:QNC5U7N7","schema_version":"1.0","canonical_sha256":"8345da7dbfc1d8dc1ec3a0ed2633376cf2cc85daf12f85c2521e3c39546dfa5b","source":{"kind":"arxiv","id":"1312.4461","version":4},"attestation_state":"computed","paper":{"title":"Low-Rank Approximations for Conditional Feedforward Computation in Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andrew Davis, Itamar Arel","submitted_at":"2013-12-16T18:58:34Z","abstract_excerpt":"Scalability properties of deep neural networks raise key research questions, particularly as the problems considered become larger and more challenging. This paper expands on the idea of conditional computation introduced by Bengio, et. al., where the nodes of a deep network are augmented by a set of gating units that determine when a node should be calculated. By factorizing the weight matrix into a low-rank approximation, an estimation of the sign of the pre-nonlinearity activation can be efficiently obtained. For networks using rectified-linear hidden units, this implies that the computatio"},"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":"1312.4461","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-12-16T18:58:34Z","cross_cats_sorted":[],"title_canon_sha256":"ca359c9f563205906c76fef62bc7fd8f5030aeb47cd54ce0e4080ffc050d4e82","abstract_canon_sha256":"fa46ed6fdcd9145726444d21caf29c766fa4f0be49ba36201f8adafbe9d0b0d5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:00:48.995007Z","signature_b64":"2XP55wB8vYfzy8XJGlU1lbC1VYteBQg5GN6orpGd2Ym/OMMR6oJ98Vh2RCZvVDDMvQgHJtU22xQ6X/u5xR2dBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8345da7dbfc1d8dc1ec3a0ed2633376cf2cc85daf12f85c2521e3c39546dfa5b","last_reissued_at":"2026-05-18T03:00:48.994365Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:00:48.994365Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Low-Rank Approximations for Conditional Feedforward Computation in Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andrew Davis, Itamar Arel","submitted_at":"2013-12-16T18:58:34Z","abstract_excerpt":"Scalability properties of deep neural networks raise key research questions, particularly as the problems considered become larger and more challenging. This paper expands on the idea of conditional computation introduced by Bengio, et. al., where the nodes of a deep network are augmented by a set of gating units that determine when a node should be calculated. By factorizing the weight matrix into a low-rank approximation, an estimation of the sign of the pre-nonlinearity activation can be efficiently obtained. For networks using rectified-linear hidden units, this implies that the computatio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1312.4461","kind":"arxiv","version":4},"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":"1312.4461","created_at":"2026-05-18T03:00:48.994432+00:00"},{"alias_kind":"arxiv_version","alias_value":"1312.4461v4","created_at":"2026-05-18T03:00:48.994432+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1312.4461","created_at":"2026-05-18T03:00:48.994432+00:00"},{"alias_kind":"pith_short_12","alias_value":"QNC5U7N7YHMN","created_at":"2026-05-18T12:27:57.521954+00:00"},{"alias_kind":"pith_short_16","alias_value":"QNC5U7N7YHMNYHWD","created_at":"2026-05-18T12:27:57.521954+00:00"},{"alias_kind":"pith_short_8","alias_value":"QNC5U7N7","created_at":"2026-05-18T12:27:57.521954+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"1701.06538","citing_title":"Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QNC5U7N7YHMNYHWDUDWSMMZXNT","json":"https://pith.science/pith/QNC5U7N7YHMNYHWDUDWSMMZXNT.json","graph_json":"https://pith.science/api/pith-number/QNC5U7N7YHMNYHWDUDWSMMZXNT/graph.json","events_json":"https://pith.science/api/pith-number/QNC5U7N7YHMNYHWDUDWSMMZXNT/events.json","paper":"https://pith.science/paper/QNC5U7N7"},"agent_actions":{"view_html":"https://pith.science/pith/QNC5U7N7YHMNYHWDUDWSMMZXNT","download_json":"https://pith.science/pith/QNC5U7N7YHMNYHWDUDWSMMZXNT.json","view_paper":"https://pith.science/paper/QNC5U7N7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1312.4461&json=true","fetch_graph":"https://pith.science/api/pith-number/QNC5U7N7YHMNYHWDUDWSMMZXNT/graph.json","fetch_events":"https://pith.science/api/pith-number/QNC5U7N7YHMNYHWDUDWSMMZXNT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QNC5U7N7YHMNYHWDUDWSMMZXNT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QNC5U7N7YHMNYHWDUDWSMMZXNT/action/storage_attestation","attest_author":"https://pith.science/pith/QNC5U7N7YHMNYHWDUDWSMMZXNT/action/author_attestation","sign_citation":"https://pith.science/pith/QNC5U7N7YHMNYHWDUDWSMMZXNT/action/citation_signature","submit_replication":"https://pith.science/pith/QNC5U7N7YHMNYHWDUDWSMMZXNT/action/replication_record"}},"created_at":"2026-05-18T03:00:48.994432+00:00","updated_at":"2026-05-18T03:00:48.994432+00:00"}