{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:FXYFWSKKEUZ3KZDLGOVK5BUU6K","short_pith_number":"pith:FXYFWSKK","canonical_record":{"source":{"id":"1807.06629","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-07-17T19:14:17Z","cross_cats_sorted":["cs.DC","cs.LG"],"title_canon_sha256":"798e08eea1ad30da92c06d20a4542745d48af1f7b2c17ec9f6568a9e0f910f47","abstract_canon_sha256":"54faa7e68eb332a41dc71cb44c35b760d32534b5c849d6493a5589fb93183f69"},"schema_version":"1.0"},"canonical_sha256":"2df05b494a2533b5646b33aaae8694f2922ab206cd247ae00ad769d5bf0e428f","source":{"kind":"arxiv","id":"1807.06629","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1807.06629","created_at":"2026-05-18T00:00:35Z"},{"alias_kind":"arxiv_version","alias_value":"1807.06629v3","created_at":"2026-05-18T00:00:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1807.06629","created_at":"2026-05-18T00:00:35Z"},{"alias_kind":"pith_short_12","alias_value":"FXYFWSKKEUZ3","created_at":"2026-05-18T12:32:25Z"},{"alias_kind":"pith_short_16","alias_value":"FXYFWSKKEUZ3KZDL","created_at":"2026-05-18T12:32:25Z"},{"alias_kind":"pith_short_8","alias_value":"FXYFWSKK","created_at":"2026-05-18T12:32:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:FXYFWSKKEUZ3KZDLGOVK5BUU6K","target":"record","payload":{"canonical_record":{"source":{"id":"1807.06629","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-07-17T19:14:17Z","cross_cats_sorted":["cs.DC","cs.LG"],"title_canon_sha256":"798e08eea1ad30da92c06d20a4542745d48af1f7b2c17ec9f6568a9e0f910f47","abstract_canon_sha256":"54faa7e68eb332a41dc71cb44c35b760d32534b5c849d6493a5589fb93183f69"},"schema_version":"1.0"},"canonical_sha256":"2df05b494a2533b5646b33aaae8694f2922ab206cd247ae00ad769d5bf0e428f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:00:35.021606Z","signature_b64":"mbOgb83auq82vxZb+Q1TkmoB4hoM4DEYeD0PjNjtXJb94vYufDX9Q555AwbVMlesUlcGkm7pIm0bMb0QaYD7DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2df05b494a2533b5646b33aaae8694f2922ab206cd247ae00ad769d5bf0e428f","last_reissued_at":"2026-05-18T00:00:35.021045Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:00:35.021045Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1807.06629","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:00:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kl9OrmRecrwVXJ0PJ+LPOIbJQfRs7LUhv3gxnziKxN5amukoMCGL1eupQqpPLXuHD7SyRK+R4wyoQdFvPdQDCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T20:28:00.083238Z"},"content_sha256":"ca548db3274db7bd750d67a53d9661e3efebdec5fe746021812acd1658ef553f","schema_version":"1.0","event_id":"sha256:ca548db3274db7bd750d67a53d9661e3efebdec5fe746021812acd1658ef553f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:FXYFWSKKEUZ3KZDLGOVK5BUU6K","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Parallel Restarted SGD with Faster Convergence and Less Communication: Demystifying Why Model Averaging Works for Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.LG"],"primary_cat":"math.OC","authors_text":"Hao Yu, Sen Yang, Shenghuo Zhu","submitted_at":"2018-07-17T19:14:17Z","abstract_excerpt":"In distributed training of deep neural networks, parallel mini-batch SGD is widely used to speed up the training process by using multiple workers. It uses multiple workers to sample local stochastic gradient in parallel, aggregates all gradients in a single server to obtain the average, and update each worker's local model using a SGD update with the averaged gradient. Ideally, parallel mini-batch SGD can achieve a linear speed-up of the training time (with respect to the number of workers) compared with SGD over a single worker. However, such linear scalability in practice is significantly l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1807.06629","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":""},"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":"2026-05-18T00:00:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3CxLxBi5fzMXxMh54UMV5/ArOKOnvJJrPDCAbu5uoew+Cex3FkdtUQnGKC0VAuwCodKSE4JerOZ0oPE9oCAgDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T20:28:00.083617Z"},"content_sha256":"0a0675baae05238c8e4e51173d08c4f0be94b4441ca08b50ca3483205898ca08","schema_version":"1.0","event_id":"sha256:0a0675baae05238c8e4e51173d08c4f0be94b4441ca08b50ca3483205898ca08"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FXYFWSKKEUZ3KZDLGOVK5BUU6K/bundle.json","state_url":"https://pith.science/pith/FXYFWSKKEUZ3KZDLGOVK5BUU6K/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FXYFWSKKEUZ3KZDLGOVK5BUU6K/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":"2026-06-03T20:28:00Z","links":{"resolver":"https://pith.science/pith/FXYFWSKKEUZ3KZDLGOVK5BUU6K","bundle":"https://pith.science/pith/FXYFWSKKEUZ3KZDLGOVK5BUU6K/bundle.json","state":"https://pith.science/pith/FXYFWSKKEUZ3KZDLGOVK5BUU6K/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FXYFWSKKEUZ3KZDLGOVK5BUU6K/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:FXYFWSKKEUZ3KZDLGOVK5BUU6K","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":"54faa7e68eb332a41dc71cb44c35b760d32534b5c849d6493a5589fb93183f69","cross_cats_sorted":["cs.DC","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-07-17T19:14:17Z","title_canon_sha256":"798e08eea1ad30da92c06d20a4542745d48af1f7b2c17ec9f6568a9e0f910f47"},"schema_version":"1.0","source":{"id":"1807.06629","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1807.06629","created_at":"2026-05-18T00:00:35Z"},{"alias_kind":"arxiv_version","alias_value":"1807.06629v3","created_at":"2026-05-18T00:00:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1807.06629","created_at":"2026-05-18T00:00:35Z"},{"alias_kind":"pith_short_12","alias_value":"FXYFWSKKEUZ3","created_at":"2026-05-18T12:32:25Z"},{"alias_kind":"pith_short_16","alias_value":"FXYFWSKKEUZ3KZDL","created_at":"2026-05-18T12:32:25Z"},{"alias_kind":"pith_short_8","alias_value":"FXYFWSKK","created_at":"2026-05-18T12:32:25Z"}],"graph_snapshots":[{"event_id":"sha256:0a0675baae05238c8e4e51173d08c4f0be94b4441ca08b50ca3483205898ca08","target":"graph","created_at":"2026-05-18T00:00:35Z","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 distributed training of deep neural networks, parallel mini-batch SGD is widely used to speed up the training process by using multiple workers. It uses multiple workers to sample local stochastic gradient in parallel, aggregates all gradients in a single server to obtain the average, and update each worker's local model using a SGD update with the averaged gradient. Ideally, parallel mini-batch SGD can achieve a linear speed-up of the training time (with respect to the number of workers) compared with SGD over a single worker. However, such linear scalability in practice is significantly l","authors_text":"Hao Yu, Sen Yang, Shenghuo Zhu","cross_cats":["cs.DC","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-07-17T19:14:17Z","title":"Parallel Restarted SGD with Faster Convergence and Less Communication: Demystifying Why Model Averaging Works for Deep Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1807.06629","kind":"arxiv","version":3},"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:ca548db3274db7bd750d67a53d9661e3efebdec5fe746021812acd1658ef553f","target":"record","created_at":"2026-05-18T00:00:35Z","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":"54faa7e68eb332a41dc71cb44c35b760d32534b5c849d6493a5589fb93183f69","cross_cats_sorted":["cs.DC","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-07-17T19:14:17Z","title_canon_sha256":"798e08eea1ad30da92c06d20a4542745d48af1f7b2c17ec9f6568a9e0f910f47"},"schema_version":"1.0","source":{"id":"1807.06629","kind":"arxiv","version":3}},"canonical_sha256":"2df05b494a2533b5646b33aaae8694f2922ab206cd247ae00ad769d5bf0e428f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2df05b494a2533b5646b33aaae8694f2922ab206cd247ae00ad769d5bf0e428f","first_computed_at":"2026-05-18T00:00:35.021045Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:00:35.021045Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mbOgb83auq82vxZb+Q1TkmoB4hoM4DEYeD0PjNjtXJb94vYufDX9Q555AwbVMlesUlcGkm7pIm0bMb0QaYD7DQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:00:35.021606Z","signed_message":"canonical_sha256_bytes"},"source_id":"1807.06629","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ca548db3274db7bd750d67a53d9661e3efebdec5fe746021812acd1658ef553f","sha256:0a0675baae05238c8e4e51173d08c4f0be94b4441ca08b50ca3483205898ca08"],"state_sha256":"7daaa34238fd95d241e0d3c1fe23ec2943ecc4fb57b80a8a16f0035e3b620291"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5hmBFUGJDBZfiizUmlNaiJr8XTW6Xzjs7HqWGam9pVfyCkjc1F90Q3uQwP+T43xk9XZycJxWilIhZQnHW9pXDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-03T20:28:00.085739Z","bundle_sha256":"130b9bc6bca99c676c090c67f0d56b2501b1f0d74fb417534ffb659e18d81c76"}}