{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:3TTR5D7ZFAS3CKVFQ26NTY6EWV","short_pith_number":"pith:3TTR5D7Z","schema_version":"1.0","canonical_sha256":"dce71e8ff92825b12aa586bcd9e3c4b574f6add55f94172bd0f080668dd89626","source":{"kind":"arxiv","id":"2110.12484","version":3},"attestation_state":"computed","paper":{"title":"Enabling Large Batch Size Training for DNN Models Beyond the Memory Limit While Maintaining Performance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"DoangJoo Synn, Jong-Kook Kim, JooYoung Park, XinYu Piao","submitted_at":"2021-10-24T16:38:05Z","abstract_excerpt":"Recent deep learning models are difficult to train using a large batch size, because commodity machines may not have enough memory to accommodate both the model and a large data batch size. The batch size is one of the hyper-parameters used in the training model, and it is dependent on and is limited by the target machine memory capacity because the batch size can only fit into the remaining memory after the model is uploaded. Moreover, the data item size is also an important factor because if each data item size is larger then the batch size that can fit into the remaining memory becomes smal"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2110.12484","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-10-24T16:38:05Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"328ce132aa763ec902e18c4c835890aa74cbaec0be6205ec97440d34d96c2783","abstract_canon_sha256":"1da89d83d2fb624524f14e968e30c129b08937bf885555d776ca13d71d9d9fdb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:38:49.978597Z","signature_b64":"+b831NtqQr7JdzMnITvL02OOSPKzerf/kAsVEw0g2JgZPBPTPQlNCzHxcZL8mEwWVp32SHYqUk1Y66B+clE0DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dce71e8ff92825b12aa586bcd9e3c4b574f6add55f94172bd0f080668dd89626","last_reissued_at":"2026-07-05T08:38:49.978171Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:38:49.978171Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enabling Large Batch Size Training for DNN Models Beyond the Memory Limit While Maintaining Performance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"DoangJoo Synn, Jong-Kook Kim, JooYoung Park, XinYu Piao","submitted_at":"2021-10-24T16:38:05Z","abstract_excerpt":"Recent deep learning models are difficult to train using a large batch size, because commodity machines may not have enough memory to accommodate both the model and a large data batch size. The batch size is one of the hyper-parameters used in the training model, and it is dependent on and is limited by the target machine memory capacity because the batch size can only fit into the remaining memory after the model is uploaded. Moreover, the data item size is also an important factor because if each data item size is larger then the batch size that can fit into the remaining memory becomes smal"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.12484","kind":"arxiv","version":3},"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/2110.12484/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2110.12484","created_at":"2026-07-05T08:38:49.978234+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.12484v3","created_at":"2026-07-05T08:38:49.978234+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.12484","created_at":"2026-07-05T08:38:49.978234+00:00"},{"alias_kind":"pith_short_12","alias_value":"3TTR5D7ZFAS3","created_at":"2026-07-05T08:38:49.978234+00:00"},{"alias_kind":"pith_short_16","alias_value":"3TTR5D7ZFAS3CKVF","created_at":"2026-07-05T08:38:49.978234+00:00"},{"alias_kind":"pith_short_8","alias_value":"3TTR5D7Z","created_at":"2026-07-05T08:38:49.978234+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3TTR5D7ZFAS3CKVFQ26NTY6EWV","json":"https://pith.science/pith/3TTR5D7ZFAS3CKVFQ26NTY6EWV.json","graph_json":"https://pith.science/api/pith-number/3TTR5D7ZFAS3CKVFQ26NTY6EWV/graph.json","events_json":"https://pith.science/api/pith-number/3TTR5D7ZFAS3CKVFQ26NTY6EWV/events.json","paper":"https://pith.science/paper/3TTR5D7Z"},"agent_actions":{"view_html":"https://pith.science/pith/3TTR5D7ZFAS3CKVFQ26NTY6EWV","download_json":"https://pith.science/pith/3TTR5D7ZFAS3CKVFQ26NTY6EWV.json","view_paper":"https://pith.science/paper/3TTR5D7Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.12484&json=true","fetch_graph":"https://pith.science/api/pith-number/3TTR5D7ZFAS3CKVFQ26NTY6EWV/graph.json","fetch_events":"https://pith.science/api/pith-number/3TTR5D7ZFAS3CKVFQ26NTY6EWV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3TTR5D7ZFAS3CKVFQ26NTY6EWV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3TTR5D7ZFAS3CKVFQ26NTY6EWV/action/storage_attestation","attest_author":"https://pith.science/pith/3TTR5D7ZFAS3CKVFQ26NTY6EWV/action/author_attestation","sign_citation":"https://pith.science/pith/3TTR5D7ZFAS3CKVFQ26NTY6EWV/action/citation_signature","submit_replication":"https://pith.science/pith/3TTR5D7ZFAS3CKVFQ26NTY6EWV/action/replication_record"}},"created_at":"2026-07-05T08:38:49.978234+00:00","updated_at":"2026-07-05T08:38:49.978234+00:00"}