{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:TQYZNWM7LTX7NZE7BQU2WBYOZF","short_pith_number":"pith:TQYZNWM7","canonical_record":{"source":{"id":"1812.06162","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-12-14T20:49:09Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"b04d8b12c826ea6cf0c514b13972734c91eb0585367f99ef5906b84d0c0df528","abstract_canon_sha256":"80c77b30385891d7316bef8c2d7a2c6fa3f2279ad8e4efc7906910fd3d019abb"},"schema_version":"1.0"},"canonical_sha256":"9c3196d99f5ceff6e49f0c29ab070ec956f92bc8f5620b730d9d9c7b29a43060","source":{"kind":"arxiv","id":"1812.06162","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1812.06162","created_at":"2026-05-17T23:58:12Z"},{"alias_kind":"arxiv_version","alias_value":"1812.06162v1","created_at":"2026-05-17T23:58:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1812.06162","created_at":"2026-05-17T23:58:12Z"},{"alias_kind":"pith_short_12","alias_value":"TQYZNWM7LTX7","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_16","alias_value":"TQYZNWM7LTX7NZE7","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_8","alias_value":"TQYZNWM7","created_at":"2026-05-18T12:32:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:TQYZNWM7LTX7NZE7BQU2WBYOZF","target":"record","payload":{"canonical_record":{"source":{"id":"1812.06162","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-12-14T20:49:09Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"b04d8b12c826ea6cf0c514b13972734c91eb0585367f99ef5906b84d0c0df528","abstract_canon_sha256":"80c77b30385891d7316bef8c2d7a2c6fa3f2279ad8e4efc7906910fd3d019abb"},"schema_version":"1.0"},"canonical_sha256":"9c3196d99f5ceff6e49f0c29ab070ec956f92bc8f5620b730d9d9c7b29a43060","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:58:12.990589Z","signature_b64":"9YFEhR2OtOVDXlk6UgpMwdlL+c87gUjucRMYFV10b36NCeWOMxm5Hm9RELoDTUM8IE8Quy9nKAViCvm1n4KpAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c3196d99f5ceff6e49f0c29ab070ec956f92bc8f5620b730d9d9c7b29a43060","last_reissued_at":"2026-05-17T23:58:12.990057Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:58:12.990057Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1812.06162","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":"2026-05-17T23:58:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zghpEHuo2z+FK7bgq8kbqG/YRw8ZtvIgRSZFCMuwoGrk9dDhP9Q5oKXVVU9F8YCF7HDRhzgonrIBYJ+HDr7/Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-08T14:44:14.235779Z"},"content_sha256":"5608d5d7eb85e79772e604f5da112237356b8131835bc32fc9af6645e2b18d13","schema_version":"1.0","event_id":"sha256:5608d5d7eb85e79772e604f5da112237356b8131835bc32fc9af6645e2b18d13"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:TQYZNWM7LTX7NZE7BQU2WBYOZF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"An Empirical Model of Large-Batch Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dario Amodei, Jared Kaplan, OpenAI Dota Team, Sam McCandlish","submitted_at":"2018-12-14T20:49:09Z","abstract_excerpt":"In an increasing number of domains it has been demonstrated that deep learning models can be trained using relatively large batch sizes without sacrificing data efficiency. However the limits of this massive data parallelism seem to differ from domain to domain, ranging from batches of tens of thousands in ImageNet to batches of millions in RL agents that play the game Dota 2. To our knowledge there is limited conceptual understanding of why these limits to batch size differ or how we might choose the correct batch size in a new domain. In this paper, we demonstrate that a simple and easy-to-m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1812.06162","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":"2026-05-17T23:58:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ElS7r1I75ST57JxGftnhZbdlwap+kIFS09czCgsPX0SiwAE5PVn6Egk2s/jTI5cRKEdvgISQEZCddELiQuZJCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-08T14:44:14.236117Z"},"content_sha256":"40f32ea12a57f66b288be1ca97770523797180233911123994cffeaa95eadb4b","schema_version":"1.0","event_id":"sha256:40f32ea12a57f66b288be1ca97770523797180233911123994cffeaa95eadb4b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TQYZNWM7LTX7NZE7BQU2WBYOZF/bundle.json","state_url":"https://pith.science/pith/TQYZNWM7LTX7NZE7BQU2WBYOZF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TQYZNWM7LTX7NZE7BQU2WBYOZF/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":"2026-06-08T14:44:14Z","links":{"resolver":"https://pith.science/pith/TQYZNWM7LTX7NZE7BQU2WBYOZF","bundle":"https://pith.science/pith/TQYZNWM7LTX7NZE7BQU2WBYOZF/bundle.json","state":"https://pith.science/pith/TQYZNWM7LTX7NZE7BQU2WBYOZF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TQYZNWM7LTX7NZE7BQU2WBYOZF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:TQYZNWM7LTX7NZE7BQU2WBYOZF","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":"80c77b30385891d7316bef8c2d7a2c6fa3f2279ad8e4efc7906910fd3d019abb","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-12-14T20:49:09Z","title_canon_sha256":"b04d8b12c826ea6cf0c514b13972734c91eb0585367f99ef5906b84d0c0df528"},"schema_version":"1.0","source":{"id":"1812.06162","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1812.06162","created_at":"2026-05-17T23:58:12Z"},{"alias_kind":"arxiv_version","alias_value":"1812.06162v1","created_at":"2026-05-17T23:58:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1812.06162","created_at":"2026-05-17T23:58:12Z"},{"alias_kind":"pith_short_12","alias_value":"TQYZNWM7LTX7","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_16","alias_value":"TQYZNWM7LTX7NZE7","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_8","alias_value":"TQYZNWM7","created_at":"2026-05-18T12:32:56Z"}],"graph_snapshots":[{"event_id":"sha256:40f32ea12a57f66b288be1ca97770523797180233911123994cffeaa95eadb4b","target":"graph","created_at":"2026-05-17T23:58:12Z","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":"In an increasing number of domains it has been demonstrated that deep learning models can be trained using relatively large batch sizes without sacrificing data efficiency. However the limits of this massive data parallelism seem to differ from domain to domain, ranging from batches of tens of thousands in ImageNet to batches of millions in RL agents that play the game Dota 2. To our knowledge there is limited conceptual understanding of why these limits to batch size differ or how we might choose the correct batch size in a new domain. In this paper, we demonstrate that a simple and easy-to-m","authors_text":"Dario Amodei, Jared Kaplan, OpenAI Dota Team, Sam McCandlish","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-12-14T20:49:09Z","title":"An Empirical Model of Large-Batch Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1812.06162","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":"sha256:5608d5d7eb85e79772e604f5da112237356b8131835bc32fc9af6645e2b18d13","target":"record","created_at":"2026-05-17T23:58:12Z","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":"80c77b30385891d7316bef8c2d7a2c6fa3f2279ad8e4efc7906910fd3d019abb","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-12-14T20:49:09Z","title_canon_sha256":"b04d8b12c826ea6cf0c514b13972734c91eb0585367f99ef5906b84d0c0df528"},"schema_version":"1.0","source":{"id":"1812.06162","kind":"arxiv","version":1}},"canonical_sha256":"9c3196d99f5ceff6e49f0c29ab070ec956f92bc8f5620b730d9d9c7b29a43060","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9c3196d99f5ceff6e49f0c29ab070ec956f92bc8f5620b730d9d9c7b29a43060","first_computed_at":"2026-05-17T23:58:12.990057Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:58:12.990057Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9YFEhR2OtOVDXlk6UgpMwdlL+c87gUjucRMYFV10b36NCeWOMxm5Hm9RELoDTUM8IE8Quy9nKAViCvm1n4KpAA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:58:12.990589Z","signed_message":"canonical_sha256_bytes"},"source_id":"1812.06162","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5608d5d7eb85e79772e604f5da112237356b8131835bc32fc9af6645e2b18d13","sha256:40f32ea12a57f66b288be1ca97770523797180233911123994cffeaa95eadb4b"],"state_sha256":"ed3d9018d5989eb78194c42ee20754c3e33b437dab3ce38ab3fff72fff6f6be5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BL4K5tmXnsG2SxC6NzKEwJ2sMxoIY5vqQVxNyR8tKMqYgtIAWDRE4fTgAJWsSCU3nYgPfuxb66z2lU2Bx5KLCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-08T14:44:14.240547Z","bundle_sha256":"5f718fa3b81d641c2a8b03ede8543cfec4f60b395c0f325b61093080f716c3bf"}}