{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:24MREQMASEPBOBGQNVSFW6KGWP","short_pith_number":"pith:24MREQMA","canonical_record":{"source":{"id":"1702.05870","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-02-20T06:17:02Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"4efe4b5cc91b8fa9cb5548727e355eefcfa62310eed93dcae4693bdce810ce29","abstract_canon_sha256":"20ecd73b68aa6e9c8be400e12d7f28b2a18dc917d78bf4713a3a418b7f3fdb52"},"schema_version":"1.0"},"canonical_sha256":"d719124180911e1704d06d645b7946b3f8d459073476dcc68f5f422d78bf1713","source":{"kind":"arxiv","id":"1702.05870","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1702.05870","created_at":"2026-05-18T00:32:20Z"},{"alias_kind":"arxiv_version","alias_value":"1702.05870v5","created_at":"2026-05-18T00:32:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1702.05870","created_at":"2026-05-18T00:32:20Z"},{"alias_kind":"pith_short_12","alias_value":"24MREQMASEPB","created_at":"2026-05-18T12:30:55Z"},{"alias_kind":"pith_short_16","alias_value":"24MREQMASEPBOBGQ","created_at":"2026-05-18T12:30:55Z"},{"alias_kind":"pith_short_8","alias_value":"24MREQMA","created_at":"2026-05-18T12:30:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:24MREQMASEPBOBGQNVSFW6KGWP","target":"record","payload":{"canonical_record":{"source":{"id":"1702.05870","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-02-20T06:17:02Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"4efe4b5cc91b8fa9cb5548727e355eefcfa62310eed93dcae4693bdce810ce29","abstract_canon_sha256":"20ecd73b68aa6e9c8be400e12d7f28b2a18dc917d78bf4713a3a418b7f3fdb52"},"schema_version":"1.0"},"canonical_sha256":"d719124180911e1704d06d645b7946b3f8d459073476dcc68f5f422d78bf1713","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:32:20.810083Z","signature_b64":"7vhhX5AZ4vWMx7Pb6kqQNYVKuGWi4RQVCTqKbOLpBg9XITuKhAGGgNRDdT/zCfqPV0sVK39SHuWof5HT5SDGBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d719124180911e1704d06d645b7946b3f8d459073476dcc68f5f422d78bf1713","last_reissued_at":"2026-05-18T00:32:20.809627Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:32:20.809627Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1702.05870","source_version":5,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:32:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Smk6rylkCaNRrtdhnUYvSStnSvRzvnwRZiOtrDb7cZgCk8Yd5eAnI/DmGivxqZYNdF3N/z8UONEsZjW4ISjyDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T21:18:27.454505Z"},"content_sha256":"f14c50c2043c85167a7bc318942dabf69522efc24a0a73c741937808b18787f4","schema_version":"1.0","event_id":"sha256:f14c50c2043c85167a7bc318942dabf69522efc24a0a73c741937808b18787f4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:24MREQMASEPBOBGQNVSFW6KGWP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Cosine Normalization: Using Cosine Similarity Instead of Dot Product in Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Chunjie Luo, Jianfeng Zhan, Lei Wang, Qiang Yang","submitted_at":"2017-02-20T06:17:02Z","abstract_excerpt":"Traditionally, multi-layer neural networks use dot product between the output vector of previous layer and the incoming weight vector as the input to activation function. The result of dot product is unbounded, thus increases the risk of large variance. Large variance of neuron makes the model sensitive to the change of input distribution, thus results in poor generalization, and aggravates the internal covariate shift which slows down the training. To bound dot product and decrease the variance, we propose to use cosine similarity or centered cosine similarity (Pearson Correlation Coefficient"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1702.05870","kind":"arxiv","version":5},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:32:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Lse/gmC0ae3rpvjAgjDjAv3r6k53BTHJYgGAkUigkv1m7BDU4J/XGWKoevbGqR9IdtmiO6Ek8u84WS0o5Eq4Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T21:18:27.454854Z"},"content_sha256":"0e8e915d8e62efb6e84028457a5dba1bfe1ec92bab3320e3221ce3993a3345b3","schema_version":"1.0","event_id":"sha256:0e8e915d8e62efb6e84028457a5dba1bfe1ec92bab3320e3221ce3993a3345b3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/24MREQMASEPBOBGQNVSFW6KGWP/bundle.json","state_url":"https://pith.science/pith/24MREQMASEPBOBGQNVSFW6KGWP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/24MREQMASEPBOBGQNVSFW6KGWP/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-03T21:18:27Z","links":{"resolver":"https://pith.science/pith/24MREQMASEPBOBGQNVSFW6KGWP","bundle":"https://pith.science/pith/24MREQMASEPBOBGQNVSFW6KGWP/bundle.json","state":"https://pith.science/pith/24MREQMASEPBOBGQNVSFW6KGWP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/24MREQMASEPBOBGQNVSFW6KGWP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:24MREQMASEPBOBGQNVSFW6KGWP","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"20ecd73b68aa6e9c8be400e12d7f28b2a18dc917d78bf4713a3a418b7f3fdb52","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-02-20T06:17:02Z","title_canon_sha256":"4efe4b5cc91b8fa9cb5548727e355eefcfa62310eed93dcae4693bdce810ce29"},"schema_version":"1.0","source":{"id":"1702.05870","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1702.05870","created_at":"2026-05-18T00:32:20Z"},{"alias_kind":"arxiv_version","alias_value":"1702.05870v5","created_at":"2026-05-18T00:32:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1702.05870","created_at":"2026-05-18T00:32:20Z"},{"alias_kind":"pith_short_12","alias_value":"24MREQMASEPB","created_at":"2026-05-18T12:30:55Z"},{"alias_kind":"pith_short_16","alias_value":"24MREQMASEPBOBGQ","created_at":"2026-05-18T12:30:55Z"},{"alias_kind":"pith_short_8","alias_value":"24MREQMA","created_at":"2026-05-18T12:30:55Z"}],"graph_snapshots":[{"event_id":"sha256:0e8e915d8e62efb6e84028457a5dba1bfe1ec92bab3320e3221ce3993a3345b3","target":"graph","created_at":"2026-05-18T00:32:20Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Traditionally, multi-layer neural networks use dot product between the output vector of previous layer and the incoming weight vector as the input to activation function. The result of dot product is unbounded, thus increases the risk of large variance. Large variance of neuron makes the model sensitive to the change of input distribution, thus results in poor generalization, and aggravates the internal covariate shift which slows down the training. To bound dot product and decrease the variance, we propose to use cosine similarity or centered cosine similarity (Pearson Correlation Coefficient","authors_text":"Chunjie Luo, Jianfeng Zhan, Lei Wang, Qiang Yang","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-02-20T06:17:02Z","title":"Cosine Normalization: Using Cosine Similarity Instead of Dot Product in Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1702.05870","kind":"arxiv","version":5},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f14c50c2043c85167a7bc318942dabf69522efc24a0a73c741937808b18787f4","target":"record","created_at":"2026-05-18T00:32:20Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"20ecd73b68aa6e9c8be400e12d7f28b2a18dc917d78bf4713a3a418b7f3fdb52","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-02-20T06:17:02Z","title_canon_sha256":"4efe4b5cc91b8fa9cb5548727e355eefcfa62310eed93dcae4693bdce810ce29"},"schema_version":"1.0","source":{"id":"1702.05870","kind":"arxiv","version":5}},"canonical_sha256":"d719124180911e1704d06d645b7946b3f8d459073476dcc68f5f422d78bf1713","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d719124180911e1704d06d645b7946b3f8d459073476dcc68f5f422d78bf1713","first_computed_at":"2026-05-18T00:32:20.809627Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:32:20.809627Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7vhhX5AZ4vWMx7Pb6kqQNYVKuGWi4RQVCTqKbOLpBg9XITuKhAGGgNRDdT/zCfqPV0sVK39SHuWof5HT5SDGBA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:32:20.810083Z","signed_message":"canonical_sha256_bytes"},"source_id":"1702.05870","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f14c50c2043c85167a7bc318942dabf69522efc24a0a73c741937808b18787f4","sha256:0e8e915d8e62efb6e84028457a5dba1bfe1ec92bab3320e3221ce3993a3345b3"],"state_sha256":"637de56d9b2c63eb51703344126e7f74eaee6c4c6a79d4e67c6dbe970222ab4c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/MOLJ+7oVdB2zCgKTabb0wvj09DMhLWVR1DR5lXlGJsAyYFGSG3FQoh01mNWKYEn1mArto5EYIeQudQMDy1WBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-03T21:18:27.456948Z","bundle_sha256":"6f27efac1dd6cb086137b721915d3332646e2c1eed0dbd2c861806026540eaef"}}