{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:GDMKE6GDP7FN2JJP5QXZBCVMRU","short_pith_number":"pith:GDMKE6GD","canonical_record":{"source":{"id":"2006.08517","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-15T16:18:05Z","cross_cats_sorted":["cs.CV","cs.DC","stat.ML"],"title_canon_sha256":"ac2474bdc60a30e08718c87c4c1295ef50d46e4162730cad3a1ed1a613230869","abstract_canon_sha256":"6489cf17e874b5d2d533b73c5f43a41ea11647ff49108151d13bb607eddf325f"},"schema_version":"1.0"},"canonical_sha256":"30d8a278c37fcadd252fec2f908aac8d116f8ebe5b973601da2d4e3ba4a523f0","source":{"kind":"arxiv","id":"2006.08517","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2006.08517","created_at":"2026-07-05T01:10:22Z"},{"alias_kind":"arxiv_version","alias_value":"2006.08517v1","created_at":"2026-07-05T01:10:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.08517","created_at":"2026-07-05T01:10:22Z"},{"alias_kind":"pith_short_12","alias_value":"GDMKE6GDP7FN","created_at":"2026-07-05T01:10:22Z"},{"alias_kind":"pith_short_16","alias_value":"GDMKE6GDP7FN2JJP","created_at":"2026-07-05T01:10:22Z"},{"alias_kind":"pith_short_8","alias_value":"GDMKE6GD","created_at":"2026-07-05T01:10:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:GDMKE6GDP7FN2JJP5QXZBCVMRU","target":"record","payload":{"canonical_record":{"source":{"id":"2006.08517","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-15T16:18:05Z","cross_cats_sorted":["cs.CV","cs.DC","stat.ML"],"title_canon_sha256":"ac2474bdc60a30e08718c87c4c1295ef50d46e4162730cad3a1ed1a613230869","abstract_canon_sha256":"6489cf17e874b5d2d533b73c5f43a41ea11647ff49108151d13bb607eddf325f"},"schema_version":"1.0"},"canonical_sha256":"30d8a278c37fcadd252fec2f908aac8d116f8ebe5b973601da2d4e3ba4a523f0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:10:22.272573Z","signature_b64":"TWhAC3kQlfHNSoRD91hLN5VQrm1q/ki/dSQjQVehoARDK/dpnRInW0pbZgqzt2xUi3UlN3qmKey1adlTqB9QDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"30d8a278c37fcadd252fec2f908aac8d116f8ebe5b973601da2d4e3ba4a523f0","last_reissued_at":"2026-07-05T01:10:22.272211Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:10:22.272211Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2006.08517","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":"2026-07-05T01:10:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kXvJI9pMA3BuAga0em9RxvwfWx/02QzgbnMK2INMXD2eb/2jFr9UkQW/tDsqvSERiaIbb/ezGRIgRkiERqhwCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T02:34:56.878003Z"},"content_sha256":"7ebf6274bcab91e15857b035fbb7cff4e335e072aaa7ef41783df055b611b7c1","schema_version":"1.0","event_id":"sha256:7ebf6274bcab91e15857b035fbb7cff4e335e072aaa7ef41783df055b611b7c1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:GDMKE6GDP7FN2JJP5QXZBCVMRU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Limit of the Batch Size","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.DC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Cho-Jui Hsieh, Huan Zhang, James Demmel, Yang You, Yuhui Wang, Zhao Zhang","submitted_at":"2020-06-15T16:18:05Z","abstract_excerpt":"Large-batch training is an efficient approach for current distributed deep learning systems. It has enabled researchers to reduce the ImageNet/ResNet-50 training from 29 hours to around 1 minute. In this paper, we focus on studying the limit of the batch size. We think it may provide a guidance to AI supercomputer and algorithm designers. We provide detailed numerical optimization instructions for step-by-step comparison. Moreover, it is important to understand the generalization and optimization performance of huge batch training. Hoffer et al. introduced \"ultra-slow diffusion\" theory to larg"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.08517","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2006.08517/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":"2026-07-05T01:10:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aN+DemeBIWnlokJ6eLiEmTi3QizDYSaD99hryUleBeS4XZ6cb00Y4pEu5fj4+9gNAphkzcR4P+zweXzUL8eMCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T02:34:56.878683Z"},"content_sha256":"e54b3e2232b11c3a3bdaba51a784954fc8c3a5cdc8520f41973896ef58843bea","schema_version":"1.0","event_id":"sha256:e54b3e2232b11c3a3bdaba51a784954fc8c3a5cdc8520f41973896ef58843bea"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GDMKE6GDP7FN2JJP5QXZBCVMRU/bundle.json","state_url":"https://pith.science/pith/GDMKE6GDP7FN2JJP5QXZBCVMRU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GDMKE6GDP7FN2JJP5QXZBCVMRU/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":"2026-08-08T02:34:56Z","links":{"resolver":"https://pith.science/pith/GDMKE6GDP7FN2JJP5QXZBCVMRU","bundle":"https://pith.science/pith/GDMKE6GDP7FN2JJP5QXZBCVMRU/bundle.json","state":"https://pith.science/pith/GDMKE6GDP7FN2JJP5QXZBCVMRU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GDMKE6GDP7FN2JJP5QXZBCVMRU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:GDMKE6GDP7FN2JJP5QXZBCVMRU","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":"6489cf17e874b5d2d533b73c5f43a41ea11647ff49108151d13bb607eddf325f","cross_cats_sorted":["cs.CV","cs.DC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-15T16:18:05Z","title_canon_sha256":"ac2474bdc60a30e08718c87c4c1295ef50d46e4162730cad3a1ed1a613230869"},"schema_version":"1.0","source":{"id":"2006.08517","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2006.08517","created_at":"2026-07-05T01:10:22Z"},{"alias_kind":"arxiv_version","alias_value":"2006.08517v1","created_at":"2026-07-05T01:10:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.08517","created_at":"2026-07-05T01:10:22Z"},{"alias_kind":"pith_short_12","alias_value":"GDMKE6GDP7FN","created_at":"2026-07-05T01:10:22Z"},{"alias_kind":"pith_short_16","alias_value":"GDMKE6GDP7FN2JJP","created_at":"2026-07-05T01:10:22Z"},{"alias_kind":"pith_short_8","alias_value":"GDMKE6GD","created_at":"2026-07-05T01:10:22Z"}],"graph_snapshots":[{"event_id":"sha256:e54b3e2232b11c3a3bdaba51a784954fc8c3a5cdc8520f41973896ef58843bea","target":"graph","created_at":"2026-07-05T01:10:22Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2006.08517/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large-batch training is an efficient approach for current distributed deep learning systems. It has enabled researchers to reduce the ImageNet/ResNet-50 training from 29 hours to around 1 minute. In this paper, we focus on studying the limit of the batch size. We think it may provide a guidance to AI supercomputer and algorithm designers. We provide detailed numerical optimization instructions for step-by-step comparison. Moreover, it is important to understand the generalization and optimization performance of huge batch training. Hoffer et al. introduced \"ultra-slow diffusion\" theory to larg","authors_text":"Cho-Jui Hsieh, Huan Zhang, James Demmel, Yang You, Yuhui Wang, Zhao Zhang","cross_cats":["cs.CV","cs.DC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-15T16:18:05Z","title":"The Limit of the Batch Size"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.08517","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":"sha256:7ebf6274bcab91e15857b035fbb7cff4e335e072aaa7ef41783df055b611b7c1","target":"record","created_at":"2026-07-05T01:10:22Z","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":"6489cf17e874b5d2d533b73c5f43a41ea11647ff49108151d13bb607eddf325f","cross_cats_sorted":["cs.CV","cs.DC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-15T16:18:05Z","title_canon_sha256":"ac2474bdc60a30e08718c87c4c1295ef50d46e4162730cad3a1ed1a613230869"},"schema_version":"1.0","source":{"id":"2006.08517","kind":"arxiv","version":1}},"canonical_sha256":"30d8a278c37fcadd252fec2f908aac8d116f8ebe5b973601da2d4e3ba4a523f0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"30d8a278c37fcadd252fec2f908aac8d116f8ebe5b973601da2d4e3ba4a523f0","first_computed_at":"2026-07-05T01:10:22.272211Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:10:22.272211Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TWhAC3kQlfHNSoRD91hLN5VQrm1q/ki/dSQjQVehoARDK/dpnRInW0pbZgqzt2xUi3UlN3qmKey1adlTqB9QDg==","signature_status":"signed_v1","signed_at":"2026-07-05T01:10:22.272573Z","signed_message":"canonical_sha256_bytes"},"source_id":"2006.08517","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7ebf6274bcab91e15857b035fbb7cff4e335e072aaa7ef41783df055b611b7c1","sha256:e54b3e2232b11c3a3bdaba51a784954fc8c3a5cdc8520f41973896ef58843bea"],"state_sha256":"07d179b0d11d08e6e8317c29d4ea78471912691d8336864fede4afbac21b41a5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IUa47t1iWWjSuBXaGCaspNxEOPJpI4cvNevWFKAwOLijHVIBd+IgsYOaqVZmylh0U9uC+wXwk1+oiGie1Y+CBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T02:34:56.882536Z","bundle_sha256":"f545db421cfe89813acd36e99f88c70f2c9c5f34f4d40152aa13a320d59244ce"}}