{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:5RBRRG54ZJUNRSIHBXYXBTZNS7","short_pith_number":"pith:5RBRRG54","canonical_record":{"source":{"id":"1811.12019","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-29T08:52:04Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"c430e5d409c9957add42d675b33f50df6128153e03d190d896bb7f0ab427923d","abstract_canon_sha256":"5a92363217a7fa8fd1fed755a4df0c5e1f100e4d6cf31c0f7d5344942f69f519"},"schema_version":"1.0"},"canonical_sha256":"ec43189bbcca68d8c9070df170cf2d97db3e4b6ac865f01d5b5c25cd392f392d","source":{"kind":"arxiv","id":"1811.12019","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.12019","created_at":"2026-05-17T23:49:50Z"},{"alias_kind":"arxiv_version","alias_value":"1811.12019v5","created_at":"2026-05-17T23:49:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.12019","created_at":"2026-05-17T23:49:50Z"},{"alias_kind":"pith_short_12","alias_value":"5RBRRG54ZJUN","created_at":"2026-05-18T12:32:08Z"},{"alias_kind":"pith_short_16","alias_value":"5RBRRG54ZJUNRSIH","created_at":"2026-05-18T12:32:08Z"},{"alias_kind":"pith_short_8","alias_value":"5RBRRG54","created_at":"2026-05-18T12:32:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:5RBRRG54ZJUNRSIHBXYXBTZNS7","target":"record","payload":{"canonical_record":{"source":{"id":"1811.12019","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-29T08:52:04Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"c430e5d409c9957add42d675b33f50df6128153e03d190d896bb7f0ab427923d","abstract_canon_sha256":"5a92363217a7fa8fd1fed755a4df0c5e1f100e4d6cf31c0f7d5344942f69f519"},"schema_version":"1.0"},"canonical_sha256":"ec43189bbcca68d8c9070df170cf2d97db3e4b6ac865f01d5b5c25cd392f392d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:49:50.498755Z","signature_b64":"sQRokOchIWO/5BUexe4tqapt9kT657J+XeAgyKb/Oib+XStONyBB/C3Xab1OyJF6vNIarDSpqbmgAZSBzhHGBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ec43189bbcca68d8c9070df170cf2d97db3e4b6ac865f01d5b5c25cd392f392d","last_reissued_at":"2026-05-17T23:49:50.498046Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:49:50.498046Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1811.12019","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":"2026-05-17T23:49:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sUzQWvoVC1w3xcsX3KfJGdc2tTC9rE6wus92DiGMUHSsyGA6CsukOVkspB+WuPsi5VI4vblTAZFosYqMNsnfBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-29T20:11:05.610271Z"},"content_sha256":"7481e75e557992e5a158cf01263f82cebf7c11bad7897ed1ba94d21fa6fe82be","schema_version":"1.0","event_id":"sha256:7481e75e557992e5a158cf01263f82cebf7c11bad7897ed1ba94d21fa6fe82be"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:5RBRRG54ZJUNRSIHBXYXBTZNS7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Large-Scale Distributed Second-Order Optimization Using Kronecker-Factored Approximate Curvature for Deep Convolutional Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Akira Naruse, Kazuki Osawa, Rio Yokota, Satoshi Matsuoka, Yohei Tsuji, Yuichiro Ueno","submitted_at":"2018-11-29T08:52:04Z","abstract_excerpt":"Large-scale distributed training of deep neural networks suffer from the generalization gap caused by the increase in the effective mini-batch size. Previous approaches try to solve this problem by varying the learning rate and batch size over epochs and layers, or some ad hoc modification of the batch normalization. We propose an alternative approach using a second-order optimization method that shows similar generalization capability to first-order methods, but converges faster and can handle larger mini-batches. To test our method on a benchmark where highly optimized first-order methods ar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.12019","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":"2026-05-17T23:49:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S02jTLWwRCQ/YkBWCHghX9zSUE3YyZ59oZkfFDLbqjv63I4EQ2SZQDfsT5TkknEeG+HvJH4TcsfbjTTvnzuOBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-29T20:11:05.610701Z"},"content_sha256":"dd6c7ae27e3921b9371128ad5b1f9cb8f19a7a479731b6a90e10a9d0a304311a","schema_version":"1.0","event_id":"sha256:dd6c7ae27e3921b9371128ad5b1f9cb8f19a7a479731b6a90e10a9d0a304311a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5RBRRG54ZJUNRSIHBXYXBTZNS7/bundle.json","state_url":"https://pith.science/pith/5RBRRG54ZJUNRSIHBXYXBTZNS7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5RBRRG54ZJUNRSIHBXYXBTZNS7/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":"2026-05-29T20:11:05Z","links":{"resolver":"https://pith.science/pith/5RBRRG54ZJUNRSIHBXYXBTZNS7","bundle":"https://pith.science/pith/5RBRRG54ZJUNRSIHBXYXBTZNS7/bundle.json","state":"https://pith.science/pith/5RBRRG54ZJUNRSIHBXYXBTZNS7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5RBRRG54ZJUNRSIHBXYXBTZNS7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:5RBRRG54ZJUNRSIHBXYXBTZNS7","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":"5a92363217a7fa8fd1fed755a4df0c5e1f100e4d6cf31c0f7d5344942f69f519","cross_cats_sorted":["cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-29T08:52:04Z","title_canon_sha256":"c430e5d409c9957add42d675b33f50df6128153e03d190d896bb7f0ab427923d"},"schema_version":"1.0","source":{"id":"1811.12019","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.12019","created_at":"2026-05-17T23:49:50Z"},{"alias_kind":"arxiv_version","alias_value":"1811.12019v5","created_at":"2026-05-17T23:49:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.12019","created_at":"2026-05-17T23:49:50Z"},{"alias_kind":"pith_short_12","alias_value":"5RBRRG54ZJUN","created_at":"2026-05-18T12:32:08Z"},{"alias_kind":"pith_short_16","alias_value":"5RBRRG54ZJUNRSIH","created_at":"2026-05-18T12:32:08Z"},{"alias_kind":"pith_short_8","alias_value":"5RBRRG54","created_at":"2026-05-18T12:32:08Z"}],"graph_snapshots":[{"event_id":"sha256:dd6c7ae27e3921b9371128ad5b1f9cb8f19a7a479731b6a90e10a9d0a304311a","target":"graph","created_at":"2026-05-17T23:49:50Z","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":"Large-scale distributed training of deep neural networks suffer from the generalization gap caused by the increase in the effective mini-batch size. Previous approaches try to solve this problem by varying the learning rate and batch size over epochs and layers, or some ad hoc modification of the batch normalization. We propose an alternative approach using a second-order optimization method that shows similar generalization capability to first-order methods, but converges faster and can handle larger mini-batches. To test our method on a benchmark where highly optimized first-order methods ar","authors_text":"Akira Naruse, Kazuki Osawa, Rio Yokota, Satoshi Matsuoka, Yohei Tsuji, Yuichiro Ueno","cross_cats":["cs.CV","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-29T08:52:04Z","title":"Large-Scale Distributed Second-Order Optimization Using Kronecker-Factored Approximate Curvature for Deep Convolutional Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.12019","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":"sha256:7481e75e557992e5a158cf01263f82cebf7c11bad7897ed1ba94d21fa6fe82be","target":"record","created_at":"2026-05-17T23:49:50Z","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":"5a92363217a7fa8fd1fed755a4df0c5e1f100e4d6cf31c0f7d5344942f69f519","cross_cats_sorted":["cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-29T08:52:04Z","title_canon_sha256":"c430e5d409c9957add42d675b33f50df6128153e03d190d896bb7f0ab427923d"},"schema_version":"1.0","source":{"id":"1811.12019","kind":"arxiv","version":5}},"canonical_sha256":"ec43189bbcca68d8c9070df170cf2d97db3e4b6ac865f01d5b5c25cd392f392d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ec43189bbcca68d8c9070df170cf2d97db3e4b6ac865f01d5b5c25cd392f392d","first_computed_at":"2026-05-17T23:49:50.498046Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:49:50.498046Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sQRokOchIWO/5BUexe4tqapt9kT657J+XeAgyKb/Oib+XStONyBB/C3Xab1OyJF6vNIarDSpqbmgAZSBzhHGBQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:49:50.498755Z","signed_message":"canonical_sha256_bytes"},"source_id":"1811.12019","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7481e75e557992e5a158cf01263f82cebf7c11bad7897ed1ba94d21fa6fe82be","sha256:dd6c7ae27e3921b9371128ad5b1f9cb8f19a7a479731b6a90e10a9d0a304311a"],"state_sha256":"15bdc596b537a4e6ed9e05cfe8dace5fa07e325db996ab6090ff4d0f820b7ffc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"55PxsnWs9H6Cb0QIBV+9kJv3mdVVl+c36JHa78sKP6b9WYEcLRZhx6REO4omLBIGRjpyJ2JPB7CuHPQgu0xDAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-29T20:11:05.612931Z","bundle_sha256":"853a83262ce888cceca429f9d90484d8ede33b6f6c603769caf4a240274a5ae0"}}