{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:QBSQ3R6MUIG52QCY6RKY3VHC6O","short_pith_number":"pith:QBSQ3R6M","canonical_record":{"source":{"id":"1903.12650","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-29T17:55:31Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"802aff10b447154bbad8f7e76774b2543954fb76fc784d9b8ef4534560846f46","abstract_canon_sha256":"0acdf818c5220b388028c679c5292ba6a65838277cbc95aa0f475d9893bd413a"},"schema_version":"1.0"},"canonical_sha256":"80650dc7cca20ddd4058f4558dd4e2f38d184521225dfe6142249f21834b9909","source":{"kind":"arxiv","id":"1903.12650","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.12650","created_at":"2026-05-17T23:49:53Z"},{"alias_kind":"arxiv_version","alias_value":"1903.12650v1","created_at":"2026-05-17T23:49:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.12650","created_at":"2026-05-17T23:49:53Z"},{"alias_kind":"pith_short_12","alias_value":"QBSQ3R6MUIG5","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_16","alias_value":"QBSQ3R6MUIG52QCY","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_8","alias_value":"QBSQ3R6M","created_at":"2026-05-18T12:33:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:QBSQ3R6MUIG52QCY6RKY3VHC6O","target":"record","payload":{"canonical_record":{"source":{"id":"1903.12650","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-29T17:55:31Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"802aff10b447154bbad8f7e76774b2543954fb76fc784d9b8ef4534560846f46","abstract_canon_sha256":"0acdf818c5220b388028c679c5292ba6a65838277cbc95aa0f475d9893bd413a"},"schema_version":"1.0"},"canonical_sha256":"80650dc7cca20ddd4058f4558dd4e2f38d184521225dfe6142249f21834b9909","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:49:53.547153Z","signature_b64":"eVfeq7ttVAbOeAFu6R7Fu5EQt3ooUAihxZa5UcT4JXa+r13z6DpVZzkXcY4w3oA4AFzOOW5cLKu0b2cMyio/Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80650dc7cca20ddd4058f4558dd4e2f38d184521225dfe6142249f21834b9909","last_reissued_at":"2026-05-17T23:49:53.546692Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:49:53.546692Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1903.12650","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":"2026-05-17T23:49:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ytBYR3m03rgvzxsZ2Us0+L5R4+QyFIGDpfv/hE+9Bd/rhiP9eSVJKCyDGfwNfRtSRqjTeloKbC8vJc9N+/zGAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T05:38:29.618048Z"},"content_sha256":"670297c29246acc84c10203375195a354b653440593711b2abdc4b462ec243c2","schema_version":"1.0","event_id":"sha256:670297c29246acc84c10203375195a354b653440593711b2abdc4b462ec243c2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:QBSQ3R6MUIG52QCY6RKY3VHC6O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Yet Another Accelerated SGD: ResNet-50 Training on ImageNet in 74.7 seconds","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Akihiko Kasagi, Akihiro Tabuchi, Atsushi Ike, Kohta Nakashima, Masafumi Yamazaki, Masahiro Miwa, Naoto Fukumoto, Takumi Honda, Tsuguchika Tabaru","submitted_at":"2019-03-29T17:55:31Z","abstract_excerpt":"There has been a strong demand for algorithms that can execute machine learning as faster as possible and the speed of deep learning has accelerated by 30 times only in the past two years. Distributed deep learning using the large mini-batch is a key technology to address the demand and is a great challenge as it is difficult to achieve high scalability on large clusters without compromising accuracy. In this paper, we introduce optimization methods which we applied to this challenge. We achieved the training time of 74.7 seconds using 2,048 GPUs on ABCI cluster applying these methods. The tra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.12650","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":"2026-05-17T23:49:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9CpVmI1fGgEO5zfLvRd5koVSymyhyGH4mFhh3PhAl7bcNjA/mvLRCaZWLSFgWUjLIHq+HvO2ImhxuKoRagIvAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T05:38:29.618756Z"},"content_sha256":"65ab15835108f7c92568da72d219351095619babc30636b54983606e524350ee","schema_version":"1.0","event_id":"sha256:65ab15835108f7c92568da72d219351095619babc30636b54983606e524350ee"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QBSQ3R6MUIG52QCY6RKY3VHC6O/bundle.json","state_url":"https://pith.science/pith/QBSQ3R6MUIG52QCY6RKY3VHC6O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QBSQ3R6MUIG52QCY6RKY3VHC6O/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":"2026-05-26T05:38:29Z","links":{"resolver":"https://pith.science/pith/QBSQ3R6MUIG52QCY6RKY3VHC6O","bundle":"https://pith.science/pith/QBSQ3R6MUIG52QCY6RKY3VHC6O/bundle.json","state":"https://pith.science/pith/QBSQ3R6MUIG52QCY6RKY3VHC6O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QBSQ3R6MUIG52QCY6RKY3VHC6O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:QBSQ3R6MUIG52QCY6RKY3VHC6O","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":"0acdf818c5220b388028c679c5292ba6a65838277cbc95aa0f475d9893bd413a","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-29T17:55:31Z","title_canon_sha256":"802aff10b447154bbad8f7e76774b2543954fb76fc784d9b8ef4534560846f46"},"schema_version":"1.0","source":{"id":"1903.12650","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.12650","created_at":"2026-05-17T23:49:53Z"},{"alias_kind":"arxiv_version","alias_value":"1903.12650v1","created_at":"2026-05-17T23:49:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.12650","created_at":"2026-05-17T23:49:53Z"},{"alias_kind":"pith_short_12","alias_value":"QBSQ3R6MUIG5","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_16","alias_value":"QBSQ3R6MUIG52QCY","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_8","alias_value":"QBSQ3R6M","created_at":"2026-05-18T12:33:27Z"}],"graph_snapshots":[{"event_id":"sha256:65ab15835108f7c92568da72d219351095619babc30636b54983606e524350ee","target":"graph","created_at":"2026-05-17T23:49:53Z","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":"There has been a strong demand for algorithms that can execute machine learning as faster as possible and the speed of deep learning has accelerated by 30 times only in the past two years. Distributed deep learning using the large mini-batch is a key technology to address the demand and is a great challenge as it is difficult to achieve high scalability on large clusters without compromising accuracy. In this paper, we introduce optimization methods which we applied to this challenge. We achieved the training time of 74.7 seconds using 2,048 GPUs on ABCI cluster applying these methods. The tra","authors_text":"Akihiko Kasagi, Akihiro Tabuchi, Atsushi Ike, Kohta Nakashima, Masafumi Yamazaki, Masahiro Miwa, Naoto Fukumoto, Takumi Honda, Tsuguchika Tabaru","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-29T17:55:31Z","title":"Yet Another Accelerated SGD: ResNet-50 Training on ImageNet in 74.7 seconds"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.12650","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":"sha256:670297c29246acc84c10203375195a354b653440593711b2abdc4b462ec243c2","target":"record","created_at":"2026-05-17T23:49:53Z","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":"0acdf818c5220b388028c679c5292ba6a65838277cbc95aa0f475d9893bd413a","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-29T17:55:31Z","title_canon_sha256":"802aff10b447154bbad8f7e76774b2543954fb76fc784d9b8ef4534560846f46"},"schema_version":"1.0","source":{"id":"1903.12650","kind":"arxiv","version":1}},"canonical_sha256":"80650dc7cca20ddd4058f4558dd4e2f38d184521225dfe6142249f21834b9909","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"80650dc7cca20ddd4058f4558dd4e2f38d184521225dfe6142249f21834b9909","first_computed_at":"2026-05-17T23:49:53.546692Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:49:53.546692Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eVfeq7ttVAbOeAFu6R7Fu5EQt3ooUAihxZa5UcT4JXa+r13z6DpVZzkXcY4w3oA4AFzOOW5cLKu0b2cMyio/Cw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:49:53.547153Z","signed_message":"canonical_sha256_bytes"},"source_id":"1903.12650","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:670297c29246acc84c10203375195a354b653440593711b2abdc4b462ec243c2","sha256:65ab15835108f7c92568da72d219351095619babc30636b54983606e524350ee"],"state_sha256":"2fe8d5a9a610e212ac3198d35fcdecb326609d12f57d812f7055cdf865956a55"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sYjUCXw3mfhoVNk/tOREiE9YCMeyUDGvIOXiMnvqArVCrQVCJekeBMHsP3kiSuFuLRDg5G5Q5HTy+P+ul3m7Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T05:38:29.622833Z","bundle_sha256":"0a220d68cc9ff9f6285a0bcf00e40e2c0eae71f0af1d040f7b05f733ea9dfb64"}}