{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2016:I7DTJFI4O25BLX6IPJW5IF5FAD","short_pith_number":"pith:I7DTJFI4","canonical_record":{"source":{"id":"1609.06870","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2016-09-22T08:47:58Z","cross_cats_sorted":[],"title_canon_sha256":"d3d5e9cc143a43aedb550f7e6a4d7da2d7e2bd0309a943071d649d828b020d50","abstract_canon_sha256":"db1afef378982c62de9a3261875d6de7386c3f25f79cb5d993cd8481bdf39e09"},"schema_version":"1.0"},"canonical_sha256":"47c734951c76ba15dfc87a6dd417a500f6e44eb434cba720a64acc4585e54688","source":{"kind":"arxiv","id":"1609.06870","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1609.06870","created_at":"2026-05-18T00:55:57Z"},{"alias_kind":"arxiv_version","alias_value":"1609.06870v4","created_at":"2026-05-18T00:55:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1609.06870","created_at":"2026-05-18T00:55:57Z"},{"alias_kind":"pith_short_12","alias_value":"I7DTJFI4O25B","created_at":"2026-05-18T12:30:22Z"},{"alias_kind":"pith_short_16","alias_value":"I7DTJFI4O25BLX6I","created_at":"2026-05-18T12:30:22Z"},{"alias_kind":"pith_short_8","alias_value":"I7DTJFI4","created_at":"2026-05-18T12:30:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2016:I7DTJFI4O25BLX6IPJW5IF5FAD","target":"record","payload":{"canonical_record":{"source":{"id":"1609.06870","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2016-09-22T08:47:58Z","cross_cats_sorted":[],"title_canon_sha256":"d3d5e9cc143a43aedb550f7e6a4d7da2d7e2bd0309a943071d649d828b020d50","abstract_canon_sha256":"db1afef378982c62de9a3261875d6de7386c3f25f79cb5d993cd8481bdf39e09"},"schema_version":"1.0"},"canonical_sha256":"47c734951c76ba15dfc87a6dd417a500f6e44eb434cba720a64acc4585e54688","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:55:57.602986Z","signature_b64":"bNlbwBwB+CVIXfbMrSiasPMW4cwGvy5mqpqlm8lOztu4VG5Ic95R+hyjOq2bApL4D3KC9ix8I25sLajJzRmxDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"47c734951c76ba15dfc87a6dd417a500f6e44eb434cba720a64acc4585e54688","last_reissued_at":"2026-05-18T00:55:57.602391Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:55:57.602391Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1609.06870","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:55:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ixWPf8dOJOvzpFaxUrbhHK3Q8SsMiHvTPDKfPocF2OZc9T4+X7yKpKR7F7mXxcRMQ4WhNJ0NbP1UUqlMbLzoDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-09T06:16:03.616619Z"},"content_sha256":"6a4a82a8875913c967dcd0f6bfa5c1894eeaf14f59267c159f7116f172c00888","schema_version":"1.0","event_id":"sha256:6a4a82a8875913c967dcd0f6bfa5c1894eeaf14f59267c159f7116f172c00888"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2016:I7DTJFI4O25BLX6IPJW5IF5FAD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Distributed Training of Deep Neural Networks: Theoretical and Practical Limits of Parallel Scalability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Franz-Josef Pfreundt, Janis Keuper","submitted_at":"2016-09-22T08:47:58Z","abstract_excerpt":"This paper presents a theoretical analysis and practical evaluation of the main bottlenecks towards a scalable distributed solution for the training of Deep Neuronal Networks (DNNs). The presented results show, that the current state of the art approach, using data-parallelized Stochastic Gradient Descent (SGD), is quickly turning into a vastly communication bound problem. In addition, we present simple but fixed theoretic constraints, preventing effective scaling of DNN training beyond only a few dozen nodes. This leads to poor scalability of DNN training in most practical scenarios."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1609.06870","kind":"arxiv","version":4},"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":"2026-05-18T00:55:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"23MHy1GW4cbXw9WzGwJFT5E7zrqpjnuYnLFWDSDmkEundDBKIrAptX71PcoJ8VaAfgTGRG778naOfO7i3IXJAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-09T06:16:03.617348Z"},"content_sha256":"038a17d280e435ead4eb0c2806fac279810fad3f9b963e1913f8fc6f93b0893b","schema_version":"1.0","event_id":"sha256:038a17d280e435ead4eb0c2806fac279810fad3f9b963e1913f8fc6f93b0893b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/I7DTJFI4O25BLX6IPJW5IF5FAD/bundle.json","state_url":"https://pith.science/pith/I7DTJFI4O25BLX6IPJW5IF5FAD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/I7DTJFI4O25BLX6IPJW5IF5FAD/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":"2026-06-09T06:16:03Z","links":{"resolver":"https://pith.science/pith/I7DTJFI4O25BLX6IPJW5IF5FAD","bundle":"https://pith.science/pith/I7DTJFI4O25BLX6IPJW5IF5FAD/bundle.json","state":"https://pith.science/pith/I7DTJFI4O25BLX6IPJW5IF5FAD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/I7DTJFI4O25BLX6IPJW5IF5FAD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2016:I7DTJFI4O25BLX6IPJW5IF5FAD","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":"db1afef378982c62de9a3261875d6de7386c3f25f79cb5d993cd8481bdf39e09","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2016-09-22T08:47:58Z","title_canon_sha256":"d3d5e9cc143a43aedb550f7e6a4d7da2d7e2bd0309a943071d649d828b020d50"},"schema_version":"1.0","source":{"id":"1609.06870","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1609.06870","created_at":"2026-05-18T00:55:57Z"},{"alias_kind":"arxiv_version","alias_value":"1609.06870v4","created_at":"2026-05-18T00:55:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1609.06870","created_at":"2026-05-18T00:55:57Z"},{"alias_kind":"pith_short_12","alias_value":"I7DTJFI4O25B","created_at":"2026-05-18T12:30:22Z"},{"alias_kind":"pith_short_16","alias_value":"I7DTJFI4O25BLX6I","created_at":"2026-05-18T12:30:22Z"},{"alias_kind":"pith_short_8","alias_value":"I7DTJFI4","created_at":"2026-05-18T12:30:22Z"}],"graph_snapshots":[{"event_id":"sha256:038a17d280e435ead4eb0c2806fac279810fad3f9b963e1913f8fc6f93b0893b","target":"graph","created_at":"2026-05-18T00:55:57Z","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":"This paper presents a theoretical analysis and practical evaluation of the main bottlenecks towards a scalable distributed solution for the training of Deep Neuronal Networks (DNNs). The presented results show, that the current state of the art approach, using data-parallelized Stochastic Gradient Descent (SGD), is quickly turning into a vastly communication bound problem. In addition, we present simple but fixed theoretic constraints, preventing effective scaling of DNN training beyond only a few dozen nodes. This leads to poor scalability of DNN training in most practical scenarios.","authors_text":"Franz-Josef Pfreundt, Janis Keuper","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2016-09-22T08:47:58Z","title":"Distributed Training of Deep Neural Networks: Theoretical and Practical Limits of Parallel Scalability"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1609.06870","kind":"arxiv","version":4},"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":"sha256:6a4a82a8875913c967dcd0f6bfa5c1894eeaf14f59267c159f7116f172c00888","target":"record","created_at":"2026-05-18T00:55:57Z","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":"db1afef378982c62de9a3261875d6de7386c3f25f79cb5d993cd8481bdf39e09","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2016-09-22T08:47:58Z","title_canon_sha256":"d3d5e9cc143a43aedb550f7e6a4d7da2d7e2bd0309a943071d649d828b020d50"},"schema_version":"1.0","source":{"id":"1609.06870","kind":"arxiv","version":4}},"canonical_sha256":"47c734951c76ba15dfc87a6dd417a500f6e44eb434cba720a64acc4585e54688","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"47c734951c76ba15dfc87a6dd417a500f6e44eb434cba720a64acc4585e54688","first_computed_at":"2026-05-18T00:55:57.602391Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:55:57.602391Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bNlbwBwB+CVIXfbMrSiasPMW4cwGvy5mqpqlm8lOztu4VG5Ic95R+hyjOq2bApL4D3KC9ix8I25sLajJzRmxDw==","signature_status":"signed_v1","signed_at":"2026-05-18T00:55:57.602986Z","signed_message":"canonical_sha256_bytes"},"source_id":"1609.06870","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6a4a82a8875913c967dcd0f6bfa5c1894eeaf14f59267c159f7116f172c00888","sha256:038a17d280e435ead4eb0c2806fac279810fad3f9b963e1913f8fc6f93b0893b"],"state_sha256":"b9d069b3a5d50600dfef63cf34282b326ba19409ffe7d37bd1f875454c5ab974"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6bH9m7F1fGabwqA/huMyUkuh72TVowps5FAGE1gaG6JWvyOs+lFL0MqgXfvbSsRxJ7e3Tcp7BOmKMQ7sT+GmCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-09T06:16:03.621203Z","bundle_sha256":"6e8547ffd1fd66a86bb3dd2fa5d6171a4fd7de03033c50965bcbeb3f32f21774"}}