{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:P7HI7IUZ2D4BWY3OX4YBCD7BZN","short_pith_number":"pith:P7HI7IUZ","canonical_record":{"source":{"id":"1811.12941","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-30T18:58:59Z","cross_cats_sorted":["cs.DC","stat.ML"],"title_canon_sha256":"4ff4ad1b988c98631cb3a9cd9fa66f8f9c1d41ca789e158250e277b4b7029b3d","abstract_canon_sha256":"81a21728a476a10bccba96774c0ca767d79a264bcc97acb58b6c77ef09e48a29"},"schema_version":"1.0"},"canonical_sha256":"7fce8fa299d0f81b636ebf30110fe1cb54e1c4eb60246b5efe9478b0e5ebed00","source":{"kind":"arxiv","id":"1811.12941","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.12941","created_at":"2026-05-17T23:59:28Z"},{"alias_kind":"arxiv_version","alias_value":"1811.12941v1","created_at":"2026-05-17T23:59:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.12941","created_at":"2026-05-17T23:59:28Z"},{"alias_kind":"pith_short_12","alias_value":"P7HI7IUZ2D4B","created_at":"2026-05-18T12:32:43Z"},{"alias_kind":"pith_short_16","alias_value":"P7HI7IUZ2D4BWY3O","created_at":"2026-05-18T12:32:43Z"},{"alias_kind":"pith_short_8","alias_value":"P7HI7IUZ","created_at":"2026-05-18T12:32:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:P7HI7IUZ2D4BWY3OX4YBCD7BZN","target":"record","payload":{"canonical_record":{"source":{"id":"1811.12941","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-30T18:58:59Z","cross_cats_sorted":["cs.DC","stat.ML"],"title_canon_sha256":"4ff4ad1b988c98631cb3a9cd9fa66f8f9c1d41ca789e158250e277b4b7029b3d","abstract_canon_sha256":"81a21728a476a10bccba96774c0ca767d79a264bcc97acb58b6c77ef09e48a29"},"schema_version":"1.0"},"canonical_sha256":"7fce8fa299d0f81b636ebf30110fe1cb54e1c4eb60246b5efe9478b0e5ebed00","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:59:28.085629Z","signature_b64":"6ZyuCtuemumm/LB34lVW5mSTAuN4wEX7HT4ZL+XEVg9CbjVBl9atdODyEk4v8KrcGoR5weVB+2ftXCDg1fbsAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7fce8fa299d0f81b636ebf30110fe1cb54e1c4eb60246b5efe9478b0e5ebed00","last_reissued_at":"2026-05-17T23:59:28.085084Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:59:28.085084Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1811.12941","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:59:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d+kNHvhjHrTjP3KUsbfDMnEPnacW915Ol/hHtPd+vLnSZxuCwKsKe1ED15bTNLMv1jyHd9jBrZmGDx0ZPd1EDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T04:52:35.567534Z"},"content_sha256":"e9c2f2246aad3d79bf45e21fc8d9ac17ed818ff1e1fb70753d91097c158f9318","schema_version":"1.0","event_id":"sha256:e9c2f2246aad3d79bf45e21fc8d9ac17ed818ff1e1fb70753d91097c158f9318"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:P7HI7IUZ2D4BWY3OX4YBCD7BZN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On the Computational Inefficiency of Large Batch Sizes for Stochastic Gradient Descent","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Amir Gholami, Joseph Gonzalez, Kai Rothauge, Michael W. Mahoney, Nikita Vemuri, Noah Golmant, Vladimir Feinberg, Zhewei Yao","submitted_at":"2018-11-30T18:58:59Z","abstract_excerpt":"Increasing the mini-batch size for stochastic gradient descent offers significant opportunities to reduce wall-clock training time, but there are a variety of theoretical and systems challenges that impede the widespread success of this technique. We investigate these issues, with an emphasis on time to convergence and total computational cost, through an extensive empirical analysis of network training across several architectures and problem domains, including image classification, image segmentation, and language modeling. Although it is common practice to increase the batch size in order t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.12941","kind":"arxiv","version":1},"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":"2026-05-17T23:59:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"opeuLdXVKwjibVSIL8ysIo8aWhtseoOBBvBaEBmeEHCQTqFKEF0v2vS4c5VI6t5qCaSy6KZA6FKh3Z/GuBlODQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T04:52:35.568238Z"},"content_sha256":"2d9b8302a1085345368db63950a2da2674cd707c58eadb507f2c6251da5dfca5","schema_version":"1.0","event_id":"sha256:2d9b8302a1085345368db63950a2da2674cd707c58eadb507f2c6251da5dfca5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/P7HI7IUZ2D4BWY3OX4YBCD7BZN/bundle.json","state_url":"https://pith.science/pith/P7HI7IUZ2D4BWY3OX4YBCD7BZN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/P7HI7IUZ2D4BWY3OX4YBCD7BZN/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":"2026-05-26T04:52:35Z","links":{"resolver":"https://pith.science/pith/P7HI7IUZ2D4BWY3OX4YBCD7BZN","bundle":"https://pith.science/pith/P7HI7IUZ2D4BWY3OX4YBCD7BZN/bundle.json","state":"https://pith.science/pith/P7HI7IUZ2D4BWY3OX4YBCD7BZN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/P7HI7IUZ2D4BWY3OX4YBCD7BZN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:P7HI7IUZ2D4BWY3OX4YBCD7BZN","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":"81a21728a476a10bccba96774c0ca767d79a264bcc97acb58b6c77ef09e48a29","cross_cats_sorted":["cs.DC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-30T18:58:59Z","title_canon_sha256":"4ff4ad1b988c98631cb3a9cd9fa66f8f9c1d41ca789e158250e277b4b7029b3d"},"schema_version":"1.0","source":{"id":"1811.12941","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.12941","created_at":"2026-05-17T23:59:28Z"},{"alias_kind":"arxiv_version","alias_value":"1811.12941v1","created_at":"2026-05-17T23:59:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.12941","created_at":"2026-05-17T23:59:28Z"},{"alias_kind":"pith_short_12","alias_value":"P7HI7IUZ2D4B","created_at":"2026-05-18T12:32:43Z"},{"alias_kind":"pith_short_16","alias_value":"P7HI7IUZ2D4BWY3O","created_at":"2026-05-18T12:32:43Z"},{"alias_kind":"pith_short_8","alias_value":"P7HI7IUZ","created_at":"2026-05-18T12:32:43Z"}],"graph_snapshots":[{"event_id":"sha256:2d9b8302a1085345368db63950a2da2674cd707c58eadb507f2c6251da5dfca5","target":"graph","created_at":"2026-05-17T23:59:28Z","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":"Increasing the mini-batch size for stochastic gradient descent offers significant opportunities to reduce wall-clock training time, but there are a variety of theoretical and systems challenges that impede the widespread success of this technique. We investigate these issues, with an emphasis on time to convergence and total computational cost, through an extensive empirical analysis of network training across several architectures and problem domains, including image classification, image segmentation, and language modeling. Although it is common practice to increase the batch size in order t","authors_text":"Amir Gholami, Joseph Gonzalez, Kai Rothauge, Michael W. Mahoney, Nikita Vemuri, Noah Golmant, Vladimir Feinberg, Zhewei Yao","cross_cats":["cs.DC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-30T18:58:59Z","title":"On the Computational Inefficiency of Large Batch Sizes for Stochastic Gradient Descent"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.12941","kind":"arxiv","version":1},"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:e9c2f2246aad3d79bf45e21fc8d9ac17ed818ff1e1fb70753d91097c158f9318","target":"record","created_at":"2026-05-17T23:59:28Z","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":"81a21728a476a10bccba96774c0ca767d79a264bcc97acb58b6c77ef09e48a29","cross_cats_sorted":["cs.DC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-30T18:58:59Z","title_canon_sha256":"4ff4ad1b988c98631cb3a9cd9fa66f8f9c1d41ca789e158250e277b4b7029b3d"},"schema_version":"1.0","source":{"id":"1811.12941","kind":"arxiv","version":1}},"canonical_sha256":"7fce8fa299d0f81b636ebf30110fe1cb54e1c4eb60246b5efe9478b0e5ebed00","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7fce8fa299d0f81b636ebf30110fe1cb54e1c4eb60246b5efe9478b0e5ebed00","first_computed_at":"2026-05-17T23:59:28.085084Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:59:28.085084Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6ZyuCtuemumm/LB34lVW5mSTAuN4wEX7HT4ZL+XEVg9CbjVBl9atdODyEk4v8KrcGoR5weVB+2ftXCDg1fbsAA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:59:28.085629Z","signed_message":"canonical_sha256_bytes"},"source_id":"1811.12941","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e9c2f2246aad3d79bf45e21fc8d9ac17ed818ff1e1fb70753d91097c158f9318","sha256:2d9b8302a1085345368db63950a2da2674cd707c58eadb507f2c6251da5dfca5"],"state_sha256":"de550e840125f483e41019b48ab3155e05ad3c6a0a3031801022f94fa5075474"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ACLGvZX5P0zbSRLDAxGHhv4BayvLCE4Qgl+o6J9VpOCzO4Jwtmc0mZxhqLEgTKOFjucO1zA/1871lXEHEj+TBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T04:52:35.572113Z","bundle_sha256":"33f4b6240b0fa4146d084346626c1abcc8841fc8bc82df83d52a9564f9fa1711"}}